Modelling wintertime sea-spray aerosols under Arctic Haze conditions

Anthropogenic and natural emissions contribute to enhanced concentrations of aerosols in the Arctic winter and early spring with most attention focusing on anthropogenic aerosols contributing to so-called Arctic Haze. Less well-studied wintertime sea-spray aerosols (SSA) under Arctic Haze conditions are the focus of this study since they can make an important contribution :: to wintertime Arctic aerosol abundances. Analysis of field campaign data also shows evidence for enhanced local sources of SSA including marine organics at Utqiaġvik ( ::::::: formerly :::::: known :: as ::::::: Barrow, :: in northern Alaska, United Statmes ::::: States) 5 during winter 2014. Models tend to underestimate sub-micron, and overestimate super-micron, SSA in the Arctic during winter, including the base version of the Weather Research Forecast, coupled with chemistry (WRF-Chem) model used here, which includes a widely used SSA source function based on Gong et al. (1997). Quasi-hemispheric simulations for winter 2014 including updated wind-speed and sea-surface temperature :::: (SST) : SSA emission dependencies, and sources of marine SSA :::::: sea-salt : organics and sea-salt-sulphate leads :::: lead to significantly improved model performance compared to observations at 10 remote Arctic sites, notably for coarse mode sodium and chloride which are reduced. The improved model also simulates more realistic contributions of SSA to inorganic aerosols at different sites, ranging from 20-93% in the observations. Two thirds of the improved model performance is from inclusion of the dependence on SSTs. Simulation of nitrate aerosols is also improved due to less heterogeneous uptake of nitric acid on SSA in the coarse mode and related increases in fine mode nitrate. This highlights the importance of interactions between natural SSA aerosols and inorganic anthropogenic aerosols 15 contributing to Arctic Haze. Simulation of organic aerosols and the fraction of sea-salt sulphate are also improved compared to observations. However, the model underestimates episodes with elevated observed concentrations of SSA components, and :::::::::: sub-micron non-sea-salt sulphate at some Arctic sites, notably at Utqiaġvik(sub-micron aerosols). : . Possible reasons are explored in higher resolution runs over northern Alaska for periods corresponding to : a : Utqiaġvik field campaign in January and

. WRF-Chem model setup. The source functions for sea-spray emissions and their main updates are summarised below. CONTROL includes only Gong et al. (1997), while HEM_NEW includes updates to the SSA emission scheme. See text for details.
Nighttime chemistry, notably heterogeneous hydrolysis of dinitrogen pentoxide leading to HNO 3 formation, is also included (Archer-Nicholls et al., 2014). The applied MOSAIC version includes secondary organic aerosol (SOA) formation from the ox-125 idation of anthropogenic and biogenic species (Shrivastava et al., 2011;Marelle et al., 2017) and is combined with SAPRC-99 gas-phase chemistry. In the base model, OA is the sum of SOA and anthropogenic emissions of organic matter (OM). Aqueous chemistry in grid-scale (Morrison, 2009) and subgrid-scale clouds (Berg et al., 2015) is also included. Aerosol sedimentation in MOSAIC is calculated throughout the atmospheric column based on the Stokes velocity scheme, as described in Marelle et al. (2017). Wet removal of aerosols by grid-resolved stratiform clouds (precipitation) includes in-cloud and below-cloud 130 removal by rain, snow, and graupel by Brownian diffusion, interception, and impaction mechanisms following Easter et al. (2004) and Chapman et al. (2009). Wet-removal due to subgrid-scale convective clouds (Berg et al., 2015) is also included in this MOSAIC version and described in previous studies Raut et al., 2017).  (NCEP, 2000)). See text for details.

Anthropogenic and natural emissions
Anthropogenic emissions are from the Evaluating the Climate and Air Quality Impacts of Short-Lived Pollutants version 135 6 (ECLIPSE v6b) inventory, with a resolution of 0.5 o x 0.5 o (Whaley et al., 2022), including emissions of organic matter (OM). Emissions of dimethyl-sulphide (DMS) and lightning nitrogen oxides (NO x ) are calculated online in the model (see Marelle et al. (2017) and references therein). Dust emissions in MOSAIC are calculated following Shaw et al. (2008). Biogenic emissions for 2014 are calculated online using Model of Emissions of Gases and Aerosol from Nature (MEGAN) model (Guenther et al., 2012). Details about the treatment of SSA emissions and their improvement in the model are provided in 140 Section 4 and summarised in Table 1.

Simulations
Two simulation domains on a polar stereo-graphic projection are used in this study, as shown in Figure 1. The first (parent) domain (d01) covers a large part of the Northern Hemisphere with 100 × 100km horizontal resolution. The boundary and initial conditions are derived from National Centres for Environmental Prediction Final meteorological reanalysis data (NCEP ment uncertainties, EBAS documentation notes uncertainties ranging between 33% and 36% for Na + , total SO 2− 4 , NO − 3 and Cl − at Alert. These high uncertainties may be related to uncertainties in the size cut-off of sub-micron filters. Uncertainties in coarse particle observations :: (d a :: < :: 10 :::: µm) : are based on the difference between high-volume (TSP) filters collected outside and 175 sub-micron filters collected inside.
Fine mode (d : a : ≤ 2.5 µm) mass concentration data from the Interagency Monitoring for Protected Visual Environments (IMPROVE) database is also used for model evaluation for Simeonof (55.3N, -160.5W), a sub-Arctic site on the Aleutians islands :::::: Aleutian ::::::: Islands, south of Alaska and an inland site, Gates of the Arctic (66.9N, -151.5W), GoA from now on, which is located 391 km south-east of Utqiaġvik town in northern Alaska (see Fig. 1). The samples are collected on-site over 24 hours 180 every three days , Malm et al. (1994) ::::::::::::::: (Malm et al., 1994)). At these two sites observations of Na + , Cl − , organic carbon (OC), NO − 3 and total SO 2− 4 are used. To compare with the OC observations at the two Alaskan sites, modelled OA is divided by 1.8, the reported ratio of OM :::::: organic :::: mass ::::: (OM)/OC in the documentation for these two stations (Malm et al., 1994). In this study, mass concentration data with d ≤ 2.5 µm are defined as fine mode aerosols, while d a < 10 µm are defined as coarse mode aerosols.

185
Sub-micron (d a < 1.0 µm) and super-micron (1.0 < d a < 10 µm) surface mass concentration data from the National Oceanic and Atmospheric Administration (NOAA) Barrow Observatory (71.3N, -156.8W), near Utqiaġvik town (Utqiaġvik from now on), is :: are : also used in this study, with daily and weekly temporal coverage, respectively. The sampling site is located 8 km northeast of Utqiaġvik town, 3km southwest of the Arctic Ocean, covered with snow during winter and 20m above mean sea level (msl), with a prevailing, east-northeast wind off the Beaufort Sea. Na + , Cl − , NO − 3 and total SO 2− 4 mass concentrations are used since observations at the other Arctic sites used in this study are not available for winter 2014, as discussed by Freud et al. (2017).
The model Stokes aerosol diameter is converted to d a using the Seinfeld and Pandis (1998) formula. Thus, the diameter of 210 modelled sub-micron particles is up to 0.73 µm (including the first four MOSAIC bins and a fraction of the 5 th bin), and supermicron particle diameters are between 0.73 to 7.3 µm (fraction 5 th bin, 6 th and 7 th bins and fraction 8 th bin). Seven MOSAIC bins and a fraction of the 8 th bin are used (modelled Stokes r d ≤ 7.3 µm) to compare with Alert and Zeppelin observations (d a < 10 µm). All model aerosol bins are used to compare with observations at Villum, where the observations are reported as TSP. For each site, modelled aerosols are estimated at the same conditions (temperature, pressure) as the reported observations. 215 Also, observed total OC is assumed to include SOA, anthropogenic OA emissions and marine organics. Thus, from now on it will be referred to as tOC, to distinguish from OA defined earlier and OM. Aerosol measurements with different size ranges (up to 1.0 µm, 2.5 µm, and 10 µm) are used to validate the model results.

Campaign data
Details about the field campaign (January 23-27 and February 24-28, 2014) measurements near Utqiaġvik, Alaska can be 220 found in KRP18 and KRP19. Briefly, atmospheric particles were collected using a rotating micro-orifice uniform deposition impactor located 2m above the snow surface at a site located 5km across the tundra from the NOAA Barrow Observatory and inland from the Arctic Ocean. The sampled particles were analysed by computer-controlled scanning electron microscopy with energy scattering X-ray spectroscopy (CCSEM-EDX) to determine the individual particle morphology and elemental composition. The analysed samples were collected either during daytime or nighttime, and only when wind directions were 225 between 75 and 225 degrees, to minimise local pollution influence. Data analysis provided information about the different chemical components as a fraction of the total number of particles sampled observed during the campaign.

Model SSA emission treatments and updates
This section introduces the treatment of SSA emissions in the base model version of WRF-Chem using the MOSAIC aerosol scheme, followed by a description of the updates to the SSA emissions implemented in the model. The model is run with 230 the original scheme (CONTROL run) and with the updates (HEM_NEW run). Results from both runs are evaluated against observations in the Arctic, in Section 5.

SSA emissions -CONTROL run
SSA emissions ::::::: emission :::::: fluxes ::: (F) in MOSAIC are calculated per particle radius r, with 1000 sub-bins per MOSAIC bin, assuming that sea-salt is a simple mix of pure NaCl and using the density function dF/dr (in particles m -2 s -1 µm -1 ) based on 235 Gong et al. (1997) (G97 from now on). The G97 source function represents the rate that seawater droplets form per unit area (sea surface) and per increase of particle radius. The fraction of Na + is calculated using the molar weight of Na + and Cl − and then the fraction of Cl − is estimated, with the total being equal to 1. The G97 density function derived from the source function is based on laboratory experiments described in Monahan et al. (1986)  where F is a function of U, the 10m-elevation wind speed, r , the particle radius at relative humidity (RH) equal to 80%, and B = (0.380−logr) 0.650 . The source function is applied for particles with dry diameters of 0.45 µm or more (equivalent to model particle diameters). For particles with dry diameters less than 0.45 µm, a correction is applied based on reported data in O'Dowd et al. (1997), since G97 overestimates the production of small particles (Gong, 2003;De Leeuw et al., 2011). G97 is based on the whitecap method, where the emission flux scales linearly with the fraction of the ocean area covered by whitecaps. Over 245 open ocean, the whitecap fraction, W(U), is determined as a function of wind speed (Monahan and Muircheartaigh (1980); MO80 from now on): This expression for W(U) is included implicitly in Equation (1) Gong et al. (1997), is still being used in global and regional models (e.g. the Community Multiscale Air Quality Modeling System (CMAQ), (Gantt et al., 2015), Goddard Earth Observing System (GEOS)-Chem, (Huang and Jaeglé, 2017), or in other models (e.g. LOTUS-EUROS) (Barthel et al., 2019) to simulate SSA, despite being relatively old. However, modelling studies have shown that G97 overestimates super-micron

Updates to SSA emissions -HEM_NEW run
Here, updates to the model treatments ::::::: treatment : of SSA emissions are described. They are included in the run HEM_NEW, which is also used as boundary conditions for the higher resolution runs over northern Alaska.

Sub-micron SSA emissions including marine organics
Previous studies have shown that there are large numbers of SSA down to 10 nm Cravigan et al., 2015;Xu et al., 2022). Also, data-based studies in the Arctic (Kirpes et al., 2019), : and over the Atlantic Ocean (O'Dowd et al., 2004;Ovadnevaite et al., 2011;Saliba et al., 2019) particles. This scheme is activated in HEM_NEW simulations. F10 is applied from the lower aerosol bin, namely 39 nm. The scheme is based on an analysis of data from a mid-latitude cruise investigating the influence of dissolved OM on the production of sub-micron SSA. The F10 SSA source function also depends on MO80 whitecap coverage and high wind speed dependence.
Organic fractions equal to 0.2 for the first and second MOSAIC bins, 0.1 for the third bin and 0.01 for the remaining bins are used following the high biogenic activity scenario which assumes high C ::::: carbon ::: (C):Chlorophyll-a (Chl-a) ratios (see Lee et al. (2010)). F11 found that higher particle organic fractions are expected in algal bloom regions with high C:Chl-a ratios and Chl-a varying between 0.4-10 µgL -1 .  Ovadnevaite et al. (2012) showed that source functions, such as Gong (2003), based on the MO80 wind speed dependence, are responsible for an overestimation of the SSA emission flux. They found a lower wind speed dependence for small particles, based on an autumn field study off the west coast of Ireland. Other factors, such as the wave field (Salisbury et al., 2013) or fetch-dependent threshold for breaking waves (Revell et al., 2019;Hartery et al., 2020), have also been shown to affect whitecap lifetime, with implications for SSA production. In a study by Salisbury et al. 290 (2014) , (SALI14 from now on), : satellite data from Quick Scatterometer (QuikSCAT) were used to derive an expression with a lower wind speed dependence compared to MO80. Here, the SALI14 parameterisation is implemented, instead of the MO80 whitecap fraction expression, since it is based on satellite data analysis providing information with global coverage including the Arctic (e.g. Chukchi Sea and Barents Sea during autumn) and south of Alaska: Based on Figure 2 in SALI14, the seasonal mean of W(U 10 ) using Eq. 3 is lower at latitudes above 40N and 40S compared to MO80 during autumn and winter.

SST dependence
Wind speed alone cannot predict SSA variability, and it is important to also include a dependence on SSTs as pointed out by, for example, data-based studies in the Arctic (Saliba et al., 2019;Liu et al., 2021b) and mid-latitudes, such as Ovadnevaite et al. (2014). Modelling studies also showed that the application of a SST dependence improves simulated SSA concentrations compared to observations (Jaeglé et al., 2011;Sofiev et al., 2011;Spada et al., 2013;Barthel et al., 2019), but not yet implemented in WRF-Chem. More specifically, previous studies tested different SSA source functions and reported that including a SST dependence improves model results, regardless of the wind speed dependence employed (Spada et al., 2013;Grythe et al., 2014;Barthel et al., 2019). However, uncertainties still remain about the influence of SSTs on SSA production (Revell 305 et al., 2019), including the role of other factors, such as seawater composition (Callaghan et al., 2014) or wave characteristics (e.g. wave speed and breaking wave type, Callaghan et al. (2012)), which might be more important than SSTs alone. Here, the JA11 SST correction factor is applied when SSTs are between -2 o C and 30 o C to evaluate the effect of SSTs on sub-and super-micron SSA emissions. SSTs are provided by the reanalyses data, in this case, FNL, and in the presence of sea-ice, SST is :::: SSTs ::: are : set equal to -1.75 o C. In that case, the SST correction factor is set to the minimum value based on Barthel et al. 310 (2019).
Average absolute differences in super-micron aerosol mass concentrations (in µgm -3 ) between HEM_NEW and CONTROL during January and February 2014 at the surface. The black x in northern Alaska shows where Utqiaġvik is located. The black circle shows Alert, Canada, the black diamond shows Villum in Greenland, while the black pentagon shows Zeppelin, Svalbard.
Total SO 2− 4 is shown. All the results are shown north of 50N. Note the different scales.

Sea-salt sulphate
A source of ss-SO 2− 4 is included in the MOSAIC SSA emission scheme (HEM_NEW), since it was not included in the base model version (CONTROL). The mass fraction of ss-SO 2− 4 is estimated to be 0.252 of the Na + mass fraction based on Kelly et al. (2010) and Neumann et al. (2016). The fraction of ss-SO 2− 4 is subtracted from the fraction of Na + , Cl − , and marine OA. Note that the total fraction of Na + , Cl − , marine OA, and ss-SO 2− 4 is equal to 1.0, and additional emissions are not added. We 320 find that, on average, the mass fraction of ss-SO 2− 4 emissions in our simulations is around 9.9% of the total SSA emissions. This can be compared with the CMAQ model where the ss-SO 2− 4 emissions are estimated to be 7% of the total SSA emissions (Kelly et al., 2010).

Evaluation of simulated wintertime SSA and other aerosols over the Arctic
First, absolute differences in simulated aerosol concentrations between the HEM_NEW and CONTROL results, averaged over 325 January and February 2014, are presented. Model results from the two runs are then evaluated against available observations of, not only Na + and Cl − , but also OA and SO 2− 4 which now include a sea-salt component, and NO − 3 which is affected by heterogeneous reactions on SSA. We also show NH + 4 for completeness. Lastly, we compare observation-based and modelled contributions of SSA to total wintertime inorganic aerosol concentrations during winter 2014.
µgm -3 , especially south of Alaska and ::: the northern Atlantic Ocean. This is due to the combined effect of using a lower wind speed dependence and including the SST dependence (Fig. 2). Inclusion of a SST dependence leads to a larger decrease in locally produced super-micron Na + and Cl − over the Arctic and sub-Arctic ice-free regions, due to lower temperatures north 335 of 50N, compared to using the lower wind speed dependence, based on SAL14, which has a smaller effect. Overall, one-third of the super-micron reductions can be attributed to the lower wind speed dependence and two-thirds to the SST dependence.
Super-micron NO − 3 is also lower (by up to 1.0 µgm -3 ) due to less formation of NO − 3 via heterogeneous uptake of HNO 3 on SSA. These reactions involving heterogeneous uptake of acid gases also produce HCl, thus depleting Cl − relative to Na + (Su et al., 2022). The presence of sea-ice also plays a role. Smaller decreases in Na + and Cl − are found north of Alaska (Beaufort SSA production over northern Alaska are examined further in Section 6. Furthermore, due to the addition of marine organics and ss-SO 2− 4 in HEM_NEW, there is more super-micron SO 2− 4 , by up to 2 µgm -3 , and super-micron OA, by up to 0.6 µgm -3 , over marine regions. Super-micron NH + 4 slightly increases up to 0.15 µgm -3 over regions where NO − 3 increases. Figure 3. ::: The :::: same :: as ::::: Figure :: 2, ::: but :: for ::::::::: sub-micron :::::: aerosol :::: mass ::::::::::: concentrations. There are smaller decreases in HEM_NEW sub-micron Na + compared to CONTROL, by up to 0.25 µgm -3 , south of Alaska 345 and in the North Atlantic (Fig. 3). Again, this is due primarily to the introduction of the SST dependence. When using SALI14 lower wind speed dependence alone, there is a small decrease in sub-micron Cl − and a small increase in sub-micron Na + over the Arctic. Sub-micron Cl − also decreases over continental areas, where NO − 3 and HNO 3 are higher due to anthropogenic sources (Fig. 3). Heterogeneous uptake on SSA reduces Cl − and increases sub-micron NO − 3 by up to 6.0 µgm -3 in HEM_NEW over continental regions while the increases over the Arctic Ocean are smaller. This is in contrast to super-micron NO − 3 350 decreases. These results are consistent with the study of Chen et al. (2016), also using WRF-Chem with MOSAIC, who noted that since SSA are primarily present in the coarse (super-micron) mode, this favours the formation of NaNO 3 which is thermodynamically stable, and limits the formation of NH 4 NO 3 which is semi-volatile . Therefore, lower in :::::: aerosol :::: mass ::::::::::: concentrations, in µgm -3 , averaged over January and February 2014 for the CONTROL and HEM_NEW simulations compared to the observations. NA stands for not available. super-micron SSA in HEM_NEW, results in less super-micron NO − 3 and more sub-micron NO − 3 . We also note that, for these reasons, sub-micron NH + 4 also increases, by up to 1.5 µgm -3 , especially over continental areas, and displays similar regional 355 patterns to sub-micron NO − 3 . Inclusion of marine organics linked to SSA, leads to increases in sub-micron OA, by up to 1.5 µgm -3 , and total SO 2− 4 increases due to the addition of ss-SO 2− 4 .

Evaluation against observations
Model results are evaluated against available observations of aerosols at different sites as shown in Figures 4, 5 and 6. These figures are grouped according to the size ranges of the measurements at the different sites as discussed in Section 3.1. Mean 360 biases and root mean square errors (RMSEs) between the observations and the model results are given in Table 2 and Table   C.1 (APPENDIX C), respectively. In the following the main findings are discussed by aerosol component.
SSA (Na + and Cl − ): Updates to the treatment of SSA emissions in HEM_NEW greatly improves modelled SSA over the Arctic with notable reductions in biases and RMSEs in Na + and Cl − compared to observations at Alert, Zeppelin (d a < 10 µm), Villum (TSP), Gates of the Arctic (GoA) (fine mode), and the sub-Arctic site, Simeonof (fine mode). Overall,

365
HEM_NEW captures the spatial variability between the observed Na + and Cl − at the different sites, in particular, the lower observed concentrations at Villum, which is surrounded by sea-ice at this time of year, and higher concentrations at Simeonof and Zeppelin. The extent to which sea-ice is present near different sites is an important factor. For example, the high variability in modelled SSA at Villum at the end of January and the middle of February 2014 is likely to be due to fluctuations in sea-ice fraction around the site (0.9-1.0 in the FNL analyses). At Utqiaġvik, the model captures super-micron Na + , whereas Cl − is 370 now underestimated due to Cl − depletion.
Sub-micron Na + and Cl − are still underestimated in HEM_NEW : at :::: this ::: site : with average biases of about -0.5 µgm -3 for Na + and -0.12 µgm -3 for Cl − with higher biases during episodes with elevated observed SSA. Sub-micron SSA at this site ::::: Here, :::::::::: sub-micron :::: SSA may have been transported to the Arctic from the Pacific Ocean (Quinn et al., 2002;May et al., 2016), and thus model underestimations may point to deficiencies in the SSA source function further south or issues related 375 either to long-range transport or to wet and dry deposition treatments in the model. However, the fact that the model agrees better with observations over the wider Arctic, as well as at sub-Arctic Simeonof, provides confidence in the modelled longrange transport as a source of Arctic (sub-micron) SSA. Simulated SSA also compares well with reported weekly averaged sub-micron Na + mass concentrations collected during January and February 2014 at Alert (0.1 µgm -3 observed, up to 0.08 µgm -3 modelled) (Leaitch et al., 2018). We also note that, at Utqiaġvik, while May et al. (2016) attributed sub-micron SSA 380 to long-range transport, KRP18 estimated that 42% of their analysed samples in the sub-micron range were fresh SSA with chemical signatures similar to sea-water, with 18% classed as partially aged with enhanced anthropogenic components (S, N) and depleted Cl − , and the remainder included organics and sulphate particles. Thus, model discrepancies may also be due to local processes influencing SSA over northern Alaska. This is investigated further in Section 6.
Nitrate: Improved SSA treatments in HEM_NEW also lead to improved simulation of NO − 3 at some sites, notably Simeonof,

385
GoA, Alert, Villum, : and biases are reduced (see Table 2). While modelled super-micron NO − 3 at Utqiaġvik is improved, the model still underestimates certain periods when elevated sub-micron NO − 3 is observed, also the case at GoA and Simeonof. The improved behaviour of modelled NO − 3 is, in general, due to reductions in Na + and Cl − , leading to less NO − 3 production in the coarse mode, especially close to or just downwind of major anthropogenic emission regions at mid-latitudes, and a shift to more NO − 3 in the fine mode, as discussed previously. These effects are most evident at Utqiaġvik, where the model can be compared 390 to sub-and super-micron data. Comparison with data from other sites is with either total, coarse plus fine mode, or fine mode aerosol observations, and therefore includes both increases and decreases in simulated NO − 3 . Overall, these results illustrate the importance of correctly simulating SSA and its effects on anthropogenic aerosols. While observed NO − 3 concentrations are generally lower than other aerosol components, such as Na + , Cl − or nss-SO 2− 4 , during Arctic winter, a recent trend analysis study showed that NO − 3 is clearly increasing at Alert, especially during the winter months (Schmale et al., 2022). Such increases 395 in NO − 3 may be due to increased NO − 3 formation due to lower acidity following SO 2 reductions, that outweigh reductions in NO x emissions at mid-latitudes (Sharma et al., 2019). However, increases in SSA over the Arctic Ocean, due to reductions in ice-covered waters, may also explain these changes (e.g. Browse et al. (2014)) although no significant trends in Na + have yet been detected (Schmale et al., 2022).  With regard to total simulated SO 2− 4 , the addition of ss-SO 2− 4 improves the model results, for example, at Simeonof where observed fine mode ss-SO 2− 4 makes a significant contribution (30-80%, up to 0.3 µgm -3 ) to total SO 2− 4 . ss-SO 2− 4 also contributes between 10-40% of total SO 2− 4 at Alert and Villum and modelled ss-SO 2− 4 agrees better with the observations. The remainder is nss-SO 2− 4 , a dominant component of Arctic Haze resulting from long-range transport from sources in Russia and Europe at these sites (Leaitch et al., 2018;Lange et al., 2018). Model results are at the lower end (up to 0.3 µgm -3 ) of reported submicron nss-SO 2− 4 mass concentrations (0.3-1.1 µgm -3 ) at Alert during winter 2014 (Leaitch et al., 2018). On the other hand, HEM_NEW further overestimates total observed SO 2− 4 at Zeppelin due to the inclusion of ss-SO 2− 4 , : especially during certain episodes with elevated concentrations. We note that Zeppelin is a mountain site at 471m, and thus discrepancies with the observations may also be due to issues simulating the vertical distribution and transport of nss-SO 2− 4 from Eurasian source regions (Hirdman et al., 2010). At Utqiaġvik, on the northern coast of Alaska, most of total observed super-micron SO 2− 4 is ss-SO 2− 4 410 (up to 0.18 µgm -3 , around 80%), and the inclusion of ss-SO 2− 4 in HEM_NEW improves agreement with the observations. With regard to total sub-micron SO 2− 4 , high mass concentrations are observed at Utqiaġvik compared to other Arctic sites, consisting mostly of nss-SO 2− 4 , peaking at 2.4 µgm -3 , much higher than total super-micron SO 2− 4 (peaking at 0.5 µgm -3 ), as also reported by Quinn et al. (2002). However, the model underestimates nss-SO 2− 4 at this site. As noted by KRP18 and KRP19, this is likely to be due to the :::: there : is :: a local influence from the North Slope of Alaska (NSA) oil fields to the east, :::: and :::: these ::::::::: emissions 415 ::: may ::: be ::::::::::::: underestimated :: in ::: the :::::: model. In a companion paper , Ioannidis et al. (2022, in prep.) the influence of these regional The lines and the symbols are the same as in Figure 4. See the text for more details. Note the different scales.

430
Ammonium: NH + 4 observations are available at all sites except for Simeonof and GoA. Observed NH + 4 concentrations are very low (below 0.2 µgm -3 ) at Alert and Villlum, with higher concentrations observed at Zeppelin. Overall there is a good agreement between the model and measurements, with very low biases and RMSEs in both runs, apart from an underestimation of elevated NH + 4 at Zeppelin. At Utqiaġvik, there is good agreement with super-micron NH + 4 , except for periods with higher observed NH + 4 (up to 0.1 µgm -3 ). However, the model underestimates periods with elevated sub-micron NH + 4 , of up to 0.4 435 µgm -3 , which is higher compared to the other sites. Temporal variations in NH + 4 during January and February 2014, generally, follow nss-SO 2− 4 as NH + 4 preferentially forms ammonium bisulfate and, to a lesser extent, ammonium sulfate in the particle phase (Schmale et al., 2022), and they have common anthropogenic origins. Previous studies also noted that NH + 4 is two times higher at Utqiaġvik than at Alert, Zeppelin and Villum, while SO 2− 4 concentrations are similar at all three stations :::: sites (Schmale et al., 2022), possibly suggesting differences in aerosol acidity at different sites. This is also found in this study 440 based on the observations and modelled ::::: model results (HEM_NEW). It is therefore interesting to investigate the effect of the improved SSA emissions on modelled aerosol acidity. For this, we estimate the neutralisation factor f, following Fisher et al.
(2011). The results are discussed in APPENDIX D. CONTROL tends to predict more acidic aerosols than observed. Based on the observations, most acidic aerosols are found at Alert, Zeppelin and Utqiaġvik (super-micron), with somewhat less acidic aerosols at Villum and Utqiaġvik (sub-micron). This is improved to some degree in HEM_NEW with aerosols becoming less 445 acidic at some sites, notably at Alert and Villum, due to decreases in simulated NO − 3 . However, modelled sub-micron aerosols at Utqiaġvik are less acidic than the observations, due to the underestimation of nss-SO 2− 4 . Overall, the updates to SSA emissions lead to somewhat less acidic anthropogenic aerosols over the Arctic, again highlighting the importance of interactions between SSA and other inorganic aerosols.
Organic aerosols: Only two sites provide tOC fine-mode observations ranging from between 0.15-0.3 µgm -3 at Simeonof 450 and 0.15and 0.5 ::: -0.5 : µgm -3 at GoA during January and February 2014. The inclusion of marine organics in HEM_NEW improves modelled OA, especially at the coastal Simeonof site. Since observations at other sites are not available for winter 2014, results are compared with other reported measurements. Shaw et al. (2010) reported sub-micron OA at Utqiaġvik equal to :::::: around 0.3 µgm -3 during winter 2008 (November to February). However, a more recent study by Moschos et al. (2022a) reported lower wintertime OA concentrations (d a < 10 µm) at this site (around 0.1 µgm -3 ), attributed mostly to primary- Overall, the model underestimates Arctic OA in common with many other models (Whaley et al., 2022). These discrepancies 460 may be due missing or underestimated anthropogenic or natural sources. For example, it is known that there are large uncertainties in anthropogenic OA emissions . The possibility of a wintertime marine OA source over northern Alaska is explored further in Section 6.

Contribution of SSA to total inorganic aerosols
Lastly, we assess the contribution of SSA to total inorganic aerosols in the Arctic during wintertime since previous studies noted 465 that they can make an important contribution to total sub-micron and super-micron mass fractions at this time of year (Quinn et al., 2002;May et al., 2016;Kirpes et al., 2018Kirpes et al., , 2019. Moschos et al. (2022b) also showed SSA dominates wintertime PM 10 (particulate matter with d a ≤ 10 µm) mass concentrations at remote Arctic sites, including Alert (56%), Baranova ( :::::: northern ::::::: Russia) :: (41%)(Russia), Utqiaġvik (66%), Villum (32%), and Zeppelin (65%). In contrast, at sites such as Tiksi (northern Russia) and Pallas (Finland), SO 2− 4 and OA dominate (70% and 55%, respectively). To investigate the contribution 470 of SSA to total mass concentrations during the period of this study, observed and modelled fractions of SSA to "total" (SSA plus inorganic) aerosols are estimated (see Table 3). It should be noted that this fraction varies between sites since not all components were measured. Taking into account the observations available at each site, the fraction of SSA to total SSA plus inorganics is higher at all the coastal sites (Utqiaġvik, Alert, Simeonof, Villum) and Zeppelin ranging from 54 to 93%.
Only at the GoA and Villum is the fraction of SSA smaller (20% and 32%, respectively). SSA fractions, calculated using the 475 HEM_NEW results, show similar patterns compared to the observations with fractions ranging between 44% and 84%. An exception, is Utqiaġvik where the modelled fraction is lower than in the observations due to low simulated sub-micron SSA concentrations. When taking into account all aerosol components in the model, including OA, BC and dust, SSA is dominant at Simeonof, Utqiaġvik (super-micron), Zeppelin and Villum (more than 54%), whereas at Alert, SSA contributes about 45%.
This analysis shows that SSA is an important fraction of total inorganic aerosols at most Arctic coastal sites during wintertime.

480
Overall, the results presented here show that the simulation of Arctic SSA, and other inorganic and organic aerosols, is improved as a result of the updated SSA emission treatments. In particular, simulated aerosols, including the coarse mode or super-micron fraction, are improved compared to the observations. The results also show that it is important to include natural SSA emissions of ss-SO 2− 4 and marine organics, although the latter are highly uncertain. Missing anthropogenic sources could also be contributing underestimation of OA and nss-SO 2− 4 . Many models in the recent AMAP model evaluation of

485
Arctic composition also showed similar discrepancies, attributed to issues with anthropogenic emissions, or model transport, deposition and aerosol formation (Whaley et al., 2022). The results presented here also confirm the importance of interactions between SSA and other inorganic aerosols via heterogeneous uptake, affecting mass concentrations and size distributions, notably NO − 3 , and thus model ability to capture wintertime Arctic Haze. =1cm Description of the regional-scale WRF-Chem model simulations at 20km resolution over northern Alaska. See text for 490 details. Simulation Name Description ALASKA_CONTROL_JAN HEM_NEW run at 20km, 23-28 January 2014 NEW_ALASKA_JAN including regional updates as in NEW_ALASKA_FEB ALASKA_CONTROL_FEB HEM_NEW run at 20km, 24-28 February 2014 LOC_ORG_FEB + Local source marine organics (Kirpes et al., 2019) SSA_WS_DEP_FEB + Sub-micron SSA wind-speed dependence (Russell et al., 2010) NEW_ALASKA_FEB + ERA5 sea-ice fraction (all regional updates) 6 Regional processes influencing SSA over northern Alaska

Local source of marine organics
The F10 parametrisation used in the 100km HEM_NEW run is based on C:Chl-a from a cruise at mid-latitudes. Whilst phy-505 toplankton blooms may not be expected in the high Arctic winter, previous studies have shown evidence of sea ice biological activity under low light conditions coupled with decreased sea ice in the Arctic (Krembs et al., 2002;Hancke et al., 2018) ::::::::::::::::::::::::::::::::::::::::::::::::::: (Krembs et al., 2002;Hancke et al., 2018). Analysis of data collected over the Arctic and North Atlantic during winter, and the winter-spring transition, also showed that the majority of sub-micron OM is highly correlated with Na + concentrations Shaw et al., 2010;Leaitch et al., 2018). More specifically, Russell observed tOC reached 0.33 µgm -3 . As mentioned previously, this discrepancy could also be due to missing local anthropogenic OA sources. Higher OA fractions in the super-micron leads to lower Na + and, as result lower NO − 3 . As indicated above in Section 5, a decrease in super-micron NO − 3 results in an increase in sub-micron NO − 3 . Sub-micron Na + increases probably due to the formation of NaNO 3 in the model. In the following runs, higher organic fractions are used instead of those from F10.
campaign and the map on the right shows the average differences between SSA_WS_DEP_FEB and LOC_ORG_FEB emission fluxes in µgm -2 s -1 . All the results are shown at the surface. Utqiaġvik is shown by the black dot. Note the different scales.
where U 18 is wind speed at 18m in ms -1 , for wind speeds between 2 and 14 ms -1 (Figure 2, RUS10). RUS10 used Na + 1 as a proxy for sub-micron NaCl, and subsequently SSA, because Na + 1 equalled sub-micron Cl − 1 on a molar basis for the 555 North Atlantic and Arctic sampling regions. Thus, Equation (5) is also used to estimate a correction factor for Cl − . Here, wind speeds in the first model layer are used, i.e. around 26m. Differences in U 18m and U 26m reach a maximum of 1 ms -1 (see Fig.E1 in APPENDIX E). Comparisons with radiosonde data at Utqiaġvik shows that the model performs well in terms of winds and temperatures (see APPENDIX E) and the role of meteorology on aerosols is not discussed further here. The correction factors are only applied to simulated number and mass of the SSA emissions when modelled wind speeds are between 2 and 14 ms -1 , 560 and when RUS10-calculated sub-micron SSA emissions are greater than model calculated SSA. In this way, SSA emissions are enhanced during periods of :::: with higher wind speeds.
To illustrate the sensitivity of the results to applying this correction, Fig. 8 shows differences in sub-micron aerosol mass concentrations compared to the run including local marine organics, as well as model SSA emission fluxes, the latter being the sum of dry mass emissions calculated in the model. The SSA emission flux is affected over ice-free model grids leading 565 to increased SSA production east and west of Utqiaġvik (by up to 0.015 µgm -2 s -1 ) while the highest increases are southwest of Alaska (by up to 0.035 µgm -2 ). This results in an increase of 0.25, 0.19 and 0.11 µgm -3 in sub-micron Na + , NO − 3 and OA, respectively, over the Utqiaġvik region and southwest Alaska during the February campaign. These results further illustrate the sensitivity of SSA emissions to wind speeds, in this case affecting fine mode aerosols. These results are in contrast to previous studies finding stronger wind speed dependencies for larger SSA particles, such as Liu et al. (2021b) who analysed aircraft 570 data, including over the Arctic. However, size dependent source functions need to be developed for the Arctic region.
Average differences between ALASKA_NEW_FEB and SSA_WS_DEP_FEB showing the effect of switching from FNL to ERA5 sea-ice fractions during the February campaign for (a) SSA emission flux (µgm -2 s -1 ), (b) sub-micron Na + and (c) super-micron Na + mass concentrations in µgm -3 . All the results are shown at the surface. Utqiaġvik is shown by the black dot.
Note the different scales.

Sea-ice fractions
The sensitivity of modelled SSA to prescribed sea-ice fractions during wintertime and the role of leads, is also investigated since KRP19 already pointed out the importance of using realistic sea-ice distributions to simulate marine aerosols. High spatial resolution images of sea-ice cover are available, including during the polar night, from a radar operating on top of a Results for February are shown in Fig. 9. The SSA emission flux (Fig. 9a) increases over a small region west of Utqiaġvik and across the North Slope of Alaska due to decreased sea-ice fraction, but decreases just to the east of Utqiaġvik and southwest 590 of Alaska due to increased sea-ice fraction. Sub-micron Na + slightly increases along the north coast of Alaska and around Utqiaġvik, by up to 0.1 µgm -3 ) (see Fig. 9b). Larger super-micron Na + are simulated by up to 0.4 µgm -3 around Utqiaġvik, and decreases by up to 0.4 µgm -3 southwest of Alaska (Fig. 9c). Higher SSA emission fluxes are simulated for February (0.035 µgm -2 s -1 ) compared to January (0.015 µgm -2 s -1 ), since there is more sea-ice in the region around Utqiaġvik and south west of Alaska in the January simulation (not shown here). Two further simulations are performed to explore model sensitivity to sea-595 ice fractions. First, ERA5 sea-ice fractions are set equal to zero to the north, west, and east of Utqiaġvik to examine the effect Average ::::::::: differences :::::: between ::::::::::::::::: ALASKA_NEW_FEB ::: and :::::::::::::::: SSA_WS_DEP_FEB ::::::: showing :: the ::::: effect of modelled aerosol composition against in-situ observations at Utqiaġvik ::::::: switching :::: from :::: FNL :: to ::::: ERA5 :::::: sea-ice ::::::: fractions : during the :::::: February :::::::: campaign ::: for of having ice-free conditions and the presence of open leads locally (as seen by the radar). Second, ERA5 sea-ice fractions are set equal to 0.75 north, west, east of Utqiaġvik and northwest of Alaska. In both cases, the model is run on a windy day (28 February 2014). The first sensitivity test leads to an increase in SSA emission fluxes by up to 0.2 µgm -2 s -1 when sea-ice fractions equal zero and to an increase of up to 1.2 µgm -3 and 0.05 µgm -3 in super-micron and sub-micron Na + , respectively.

Evaluation against observations in northern Alaska
Results from runs at 20km with and without the main changes included in the sensitivity tests (local source of marine organics, higher wind speed dependence and ERA-5 sea-ice fractions) are compared to sub-micron aerosol observations at Utqiaġvik for 610 both the January and February 2014 campaign :::::: periods (see Fig. 10). We note that there are no super-micron observations during the simulation periods ::::::: available due to the weekly sampling frequency. It is interesting to compare to these periods since the observations show different behaviours. While observed sub-micron Na + and Cl − concentrations reached up to 2.5 µgm -3 in February, they did not exceed 1 µgm -3 during January. As noted earlier such high concentrations were not observed at Alert and Villum during January and February 2014. The January run, including all the updates (ALASKA_NEW_JAN), captures better 615 sub-micron Na + and Cl − (reduced biases and RMSEs -see APPENDIX F, Table F1) although it underestimates observations by up to 0.6 and 0.8 µgm -3 (Fig. 10a), respectively, while sub-micron NO − 3 is slightly overestimated. Biases and RMSEs (see APPENDIX F, Table F2) are also slightly improved for February although sub-micron Na + , Cl − are still underestimated by up to 1.5 µgm -3 . Slightly more NO − 3 is simulated in January, even if the model still underestimates the elevated observations. Up to six times more OA is simulated during both periods in better agreement with reported observations (Moschos et al., 2022b) : , 620 as discussed in section 5.2. The sensitivity tests discussed in this section do not directly address the model underestimation of elevated episodes of sub-micron total SO 2− 4 , since it is mostly nss-SO 2− 4 , and thus the changes are small (during both simulation periods). We note that these runs at 20km are for periods including elevated aerosols and are thus more challenging for the model to reproduce. Runs at higher resolution may be needed to better resolve, for example, sea-ice distributions.
Observations of particle number concentration are also used to validate the regional model results at Utqiaġvik (see Fig. 11).

640
Overall the results the results presented so far in this Section show that modelled sub-micron SSA (Na + , Cl − ), and as a consequence NO − 3 , are more sensitive to using a higher wind speed dependence than sea-ice fractions, over northern Alaska, based on estimated biases and RMSEs for each test simulation (not shown here). Sea-ice fractions have a greater effect on super-micron SSA mass concentrations. Modelled sub-micron OA are more sensitive to a higher wind speed dependence and, to a lesser extent, the introduction of an additional source of local marine organics. However, the latter is highly uncertain. 645 6.5 Are blowing snow and/or frost flowers a source of sub-micron SSA during wintertime at Utqiaġvik?
Lastly, we consider whether enhanced SSA, in particular in the sub-micron size range, at Utqiaġvik could be due to blowing snow or frost flower sources. As noted earlier, KRP19 found no evidence of blowing snow or frost flowers at this site but that SSA originated from open leads during wintertime. Their findings are supported by the earlier laboratory study of Roscoe et al. (2011) who reported that frost flowers are not an efficient source of SSA. However, Shaw et al. (2010) found that during winter 650 at Utqiaġvik surface frost flowers forming on sea and lake ice are a source of ocean-derived :::::::::::: marine-derived : OM. Modelling studies including a source of blowing snow and frost flowers suggest that they are contributing to SSA at this time of year at Utqiaġvik, Alert and Zeppelin (Xu et al., 2013(Xu et al., , 2016Huang and Jaeglé, 2017;Rhodes et al., 2017). Here, depletion factors are calculated using modelled and observed sub-micron aerosol mass concentrations during the campaign periods.
The average values of modelled and observed DFs are shown in Table 5. Total SO 2− 4 is enriched relative to Na + in both the observations and the model results during both campaign periods, in contrast to FR20 who reported substantial depletion. In February, observed and model results both indicate SO 2− 4 enrichment relative to seawater, whereas in January, model results 675 are less enrichment ::::::: enriched compared to the observations, possibly due to underestimation of nss-SO 2− 4 . In January, observed total SO 2− 4 concentrations are 7.56 times higher than in reference seawater, possibly due to internal mixing with anthropogenic nss-SO 2− 4 , as noted by KRP18. Modelled total SO 2− 4 is less enriched than the observations (0.94 times higher than in reference seawater), likely due the model underestimation of nss-SO 2− 4 . FR20 did report a case of enrichment due to possible contamination from the ship, an anthropogenic source. The Na + depletion factor also shows enrichment during both campaigns, albeit 680 more negligible in the observations than in the model. Observed Na + depletion relative to Cl − is 1.09 or 1.19 times more than in reference seawater, during January and February, respectively. Our analysis suggests that blowing snow and frost flowers are not a significant source of SSA, at least during the campaign . ::: this :::::::: campaign :: in :::::: winter ::::: 2014.
=1cm Average modelled and observed molar ratios for sub-micron SSA, following Kirpes et al. (2019), during the campaign in January and February 2014 at Utqiaġvik. Model results from ALASKA_NEW_JAN and ALASKA_NEW_FEB simulations are used. Observations refer to sub-micron data from NOAA. Molar ratios Model Observations total-SO 2− 4 :Na + 0.12 0.55 Cl − :Na + 0.71 1.1 total-SO 2− 4 :Na + 1.5 0.2 Cl − :Na + 0.8 1.08 SSA can also play an important role in polar tropospheric ozone :: O 3 and halogen chemistry through the release of active bromine during spring (Fan and Jacob, 1992;Simpson et al., 2007;Peterson et al., 2017). Reactions involving bromine are an important sink of ozone (O 3 ) ::: O 3 (e.g. (Barrie, 1986;Marelle et al., 2021)). Br − depletion relative to Na + is calculated 690 only during February since observed Br − was zero during the January campaign period. The results for February show a small depletion indicating a seawater origin. The observed mass ratio of Br − to Na + ranges between 0.0057 and 0.0059, while the mass ratio of Br − to Na + in reference seawater is equal to 0.006. FR20 reported no or little Br − depletion relative to Na + due to Br − losses at the surface and small depletion further aloft (in Antarctica). For a more comprehensive analysis, observations are required at different locations and altitudes across northern Alaska. We note that the version of WRF-Chem used in this 695 study does not include halogen chemistry. It has since been implemented in a later version by Marelle et al. (2021) to examine springtime ozone depletion events at Utqiaġvik. Heterogeneous reactions on SSA from the sublimation of lofted blowing snow were :::: also included. Their results suggested that blowing snow could be a source of SSA during spring although it should be noted that this model version, including blowing snow as a source of SSA, overestimated SSA (d a < 10 µm) at Arctic sites, such as Alert and Villum during spring and did not examine wintertime conditions.
KRP18 reported the presence of both fresh (locally-produced) SSA and aged (partially Cl − -depleted) SSA for sub-micron SSA, while super-micron were mostly fresh (KRP18, Figure 2). Based on the analysis above (including observations), there is little evidence suggesting that blowing snow or frost flowers are a significant source of SSA during the campaign at Utqiaġvik 720 ::::: during :::::: winter, : and open leads are an important primary source, in agreement with KRP19.

Conclusions
In this study, the ability of the WRF-Chem model to simulate wintertime Arctic aerosols is assessed with a particular focus on SSA under Arctic Haze conditions. The inclusion of updated treatments of SSA emissions leads to improved simulation of SSA over the wider Arctic compared to the still widely used Gong et al. (1997)-based source function included in the base 725 model. Na + and Cl − biases are reduced by a factor of 7 to 16 compared to observations at Alert, Villum and Zeppelin, and by a factor of 4 compared to super-micron Na + and Cl − data at Utqiaġvik. The addition of the : a SST dependence has a larger effect on modelled SSA compared to updating the wind speed dependence, and is responsible for two-thirds of the reductions in super-micron/coarse mode SSA, due to low SSTs in the Arctic. The use of a more realistic lower wind speed dependence, based on satellite data, also results in lower super-micron SSA, but up to 5 times less compared to the SST dependence. In 730 addition to uncertainties in wind speed and SST dependencies influencing SSA production ::::::::: production :: of :::: SSA, other factors such as seawater composition, wave characteristics, fetch and salinity may also be playing a role and should be considered in future versions of WRF-Chem. Other SST dependencies could also be tested which could increase sub-micron SSA at low temperatures (Sofiev et al., 2011;Salter et al., 2015;Barthel et al., 2019). In addition, missing sources of ultrafine SSA particles, for example, due to breaking waves at the coast, could ::: also be included by defining the a : surf zone in the model 735 (Clarke et al., 2006). In all cases, more field data is needed to understand and develop SSA source functions ::::: source :::::::: functions :: for ::::: SSA : specific to the Arctic during winter.
Results from this study also highlight the importance of interactions between SSA and other inorganic aerosols, notably NO − 3 , which have largely anthropogenic origins, and contribute to wintertime Arctic Haze. Improved simulation of Na + and Cl − leads to less coarse mode and more fine mode NO − 3 in the model, in better agreement with the observations. This is due 740 to less formation of NO − 3 via heterogeneous uptake of HNO 3 , primarily in the coarse mode, and more NO − 3 in the fine mode, in line with previous mid-latitude studies. As a result, simulated aerosols in the updated model are slightly less acidic in the Arctic, improving agreement with some Arctic sites, even if the model tends to have ::::::: simulate : aerosols, which are too acidic (at some sites).
Marine organic aerosol and sub-micron sea-spray emissions :::::: aerosols : are also activated in the model since they are an important component of SSA in the Arctic, and globally, and a source of ss-SO 2− 4 is also added. Simulated OA is improved at the Simeonof sub-Arctic site with reduced biases, by up to a factor of 4, although, in general, OA is underestimated at sites over the wider Arctic. The addition of ss-SO 2− 4 agrees well with ss-SO 2− 4 derived from the observations at most Arctic sites and leads to improved modelled total SO 2− 4 . However, at Zeppelin and Villum, which are dominated by nss-SO 2− 4 , this additional source results in further overestimation. While super-micron SO 2− 4 , primarily of sea-salt origin, is captured better at 750 Utqiaġvik on the northern Alaskan coast, sub-micron SO 2− 4 , which is primarily nss-SO 2− 4 , is underestimated at this site during episodes with elevated concentrations, and also at Gates of the Arctic :::: GoA further inland. Model discrepancies in OA and nss-SO 2− 4 may be due to missing local anthropogenic emissions, coupled with missing heterogeneous or dark reactions leading to secondary aerosol formation. In the case of OA, primary marine emissions may also be underestimated. It can be noted that such underestimations are a common feature in other models (Whaley et al., 2022). Uncertainties ::: The :::::: above, ::::::::: combined :::: with 755 :::::::::: uncertainties : in model transport and wet and dry deposition processes may also be responsible for deficiencies in modelled :::::::: contribute :: to :::::: model :::::::::: deficiencies :: in ::::::::: simulating wintertime Arctic aerosols (Whaley et al., 2022).
Model sensitivity to different processes affecting wintertime SSA over northern Alaska is explored further with the aim to understand, in particular, model underestimation of sub-micron SSA at Utqiaġvik during winter 2014 when field data analysis showed that marine emissions from open leads were an important source of SSA, including marine organics (KRP18, KRP19).

760
Based on observed ratios of OC:Na + from the KRP19 campaign, a local source of marine organics is included in model runs at 20km over northern Alaska. This results in higher modelled OA, in better agreement with previous measurements at this site, and other sites such as Alert, although the model still tends to underestimate reported data.
The sensitivity of modelled SSA over northern Alaska to using a higher wind speed dependence, based on Arctic data, is also investigated. This leads to an increase in modelled sub-micron SSA, with the model performing better during January than in 765 February. Model sensitivity to prescribed sea-ice fractions is also explored. In a run with ERA5 instead of FNL sea-ice fractions, modelled super-micron SSA are more sensitive to sea-ice treatments than sub-micron SSA. In general, modelled sub-micron SSA are more sensitive to the use of a higher wind speed dependence rather than the distribution of sea-ice. To improve model simulations in this region, field campaigns are needed to study processes influencing wintertime SSA production ::::::::: production :: of :::: SSA and to determine more realistic sea-ice fractions which vary on at least a daily basis. The use of satellite sea-ice data, 770 combined with higher resolution simulations over Utqiaġvik, will also help to gain further insights into the influence of open leads on SSA production ::::::::: production :: of :::: SSA, including marine organics, during wintertime.
Missing local anthropogenic sources could also explain some of the discrepancies in modelled sub-micron SSA. For example, anthropogenic sources of Cl − may need to be considered, such as coal combustion, waste incineration, and other industrial activities (Wang et al., 2019) which are not included in current emission inventories. WRF-Chem, and models in general, also 775 lack anthropogenic emissions of Na + , which could possibly account for up to 30% of Na + , as noted by Barrie and Barrie (1990). However, the analysis of depletion factors and molar ratios, presented here for Utqiaġvik, shows that the main source of fresh SSA is from marine sources including open ocean or leads. We also find no evidence for frost flowers or blowing snow as a source of SSA at Utqiaġvik, in agreement with the findings of KRP19 and previous studies (May et al., 2016). Further insights into wintertime marine SSA sources, including organics are needed, as well as improved quantification of local 780 anthropogenic emissions.
Overall, we find that wintertime SSA at remote Arctic sites contribute between 54% and 84% to total inorganic SSA (observations and improved model results), in agreement with previous findings, that SSA are important contributors to super-micron (coarse mode, TSP) mass concentrations. Ice fractures and the area of open ocean are likely to become more important with decreasing sea-ice cover in the Arctic as a result of climate warming. This may lead to more SSA which can act as CCN or INPs 785 with implications for Arctic aerosol-cloud indirect effects :::::::: feedbacks, notably long-wave radiative forcing which dominates ::::: effects :::::: which :::::::: dominate : in winter (Eidhammer et al., 2010;Partanen et al., 2014). As well as ground-based measurements, vertical profiles of SSA components are also needed to better understand SSA sources and their impacts on clouds. Such studies will ultimately help to reduce uncertainties in estimates of aerosol-cloud indirect radiative effects and the magnitude of the associated radiative cooling (Horowitz et al., 2020) or warming (Zhao and Garrett, 2015).  (2018)). Table A1. ::::::::: WRF-Chem Land Surface model's ::::: surface : (NOAH MP) parametrization ::::::::::: parametrisation :::::: scheme. "Opt_" indicates the namelist option for NOAH MP.

Appendix C
In this APPENDIX, the ::: The : RMSEs are shown for each site and aerosol component at 100km.

Appendix D
To investigate aerosol acidity, the mean neutralized factor (f) is calculated as the ratio of NH + 4 to the sum of (2nss-SO 2− 4 + NO − 3 ), in molar concentrations, following Fisher et al. (2011), for sites in the Arctic with available observations of these aerosols. When f is equal to 1 then aerosols are assumed to be neutralized, while when f < 1 then aerosols are acidic, and more acidic when f is closer to zero (Fisher et al., 2011). In general, higher molar concentrations were observed for nss-SO 2− 4 825 compared to NO − 3 and NH + 4 . Table D1 shows f for observations and the two 100km simulations at the different sites. Overall, modelled f increases due to the improved treatment of SSA and the associated influence on NO − 3 via heterogeneous reactions. Since aerosols are assumed to be internally mixed in the model, NH + 4 and nss-SO 2− 4 mass concentrations also vary between the two simulations. Thus, aerosols in HEM_NEW tend to be less acidic (e.g. at Alert and Villum), due to NO − 3 decreases in the coarse-mode/TSP size range. This leads to better agreement with the observed f at Alert, in particular. At Villum, 830 observed aerosols are less acidic than in the model. This could be due to the fact that the model has more NH + 4 compared to the observations. Only small changes are found at Utqiaġvik between the two runs, and the model tends to have aerosols which are slightly more acidic (super-micron) and less acidic (sub-micron) compared to the observations. The small increase in model sub-micron f at Utqiaġvik could be due to the increase in sub-micron NO − 3 and insignificant changes in NH + 4 and nss-SO 2− 4 . Differences with the observed values could be explained by underestimation of nss-SO 2− 4 at this site. The calculated 835 f for observations could also be biased low (too acidic), since some of the NO − 3 and SO 2− 4 are present as Na 2 SO 4 and NaNO 3 in the atmosphere, which are not measured. to evaluate the model's performance at different altitudes. Radiosonde data (every 12h) are derived from Integrated Global Radiosonde Archive version 2 (IGRA 2) (Durre et al. (2018)). Site :::::::: Utqiaġvik :::: site is located at latitude: 71.2 :: N and longitude: -156.7. In the future, more detailed studies it is advisable to investigate the meteorological and synoptic conditions that occur at each site within Arctic. A detailed focus on the meteorology and removal treatments at lower latitudes to minimize potential 845 uncertainties linked to transport errors in the model is also desirable. ::: W. Figure E2. Time series of observed and modelled 2m and 10m temperature ::::::::: temperatures, and 10m wind speed ::::: speeds, at Barrow ::::::: Utqiaġvik, Alaska, in UTC. The observations are shown in red and derived :: are : from the NOAA observatory ::::::::: Observatory. The blue line shows the results for the HEM_NEW simulation at 100km, while the black line shows the :::: lines :::: show : results for ALASKA_NEW_JAN and ALASKA_NEW_FEB simulations at 20km. The observations are hourly, while the model output is every 3h. Table E1. Biases and RMSEs, in ::::::::: temperature :: ( o C : ), ::: wind ::::: speed :: (ms -1 : ) : and ::: wind ::::::::: directions : (degrees), are calculated between ALASKA_NEW_JAN, ALASKA_NEW_FEB and in-situ meteorological parameters derived ::::::::: observations : from ::: the NOAA Baseline Observatories :::::::: Observatory : during the campaign 's periods in January and February 2014. Bias was :::::: Absolute ::::: biases ::: are : calculated as the difference between model simulation :::::::: simulations : and :: the : observations.  Table F1. Biases :::::: Absolute ::::: biases : and RMSEs, in ::::: aerosol ::::: mass ::::::::::: concentrations :: (in : µgm -3 ,are calculated for aerosols : ) at Utqiaġvik, north of Alaska, during January 2014 and for ALASKA_CONTROL_JAN and ALASKA_NEW_JAN simulations at 20km : , ::::::: compared :: to ::::::::: observations.  Table F2. Biases :::::: Absolute ::::: biases : and RMSEs, in ::::: aerosol ::::: mass ::::::::::: concentrations :: (in : µgm -3 , calculated for aerosols : ) : at Utqiaġvik, north of Alaska, during February 2014 and for ALASKA_CONTROL_FEB and ALASKA_NEW_FEB simulations at 20km : , ::::::: compared ::: to ::::::::: observations. Author contributions. The first author (EI) implemented the updates, performed the simulations and the analysis, and drafted the paper. KSL designed the study, contributed to the interpretation of results and the analysis, and to writing the paper. JCR, LM and TO contributed to discussions about the model setup, simulations. JCR, LM, KP, RMK, PKQ and LU contributed to the analysis and interpretation of the results.

ALASKA_CONTROL_FEB
AM and HS contributed to the interpretation of the results. TT and AW contributed to the interpretation of SMPS analysis. All co-authors contributed to the paper and the discussions about the results, as well as for the revision of the manuscript and have approved the final version.
Competing interests. The authors declare that they have no conflict of interest.