Sea ice and pollution-modulated changes in Greenland ice core methanesulfonate and bromine

Abstract. Reconstruction of past changes in Arctic sea ice extent may be critical for understanding its future evolution. Methanesulfonate (MSA) and bromine concentrations preserved in ice cores have both been proposed as indicators of past sea ice conditions. In this study, two ice cores from central and north-eastern Greenland were analysed at sub-annual resolution for MSA (CH3SO3H) and bromine, covering the time period 1750–2010. We examine correlations between ice core MSA and the HadISST1 ICE sea ice dataset and consult back trajectories to infer the likely source regions. A strong correlation between the low-frequency MSA and bromine records during pre-industrial times indicates that both chemical species are likely linked to processes occurring on or near sea ice in the same source regions. The positive correlation between ice core MSA and bromine persists until the mid-20th century, when the acidity of Greenland ice begins to increase markedly due to increased fossil fuel emissions. After that time, MSA levels decrease as a result of declining sea ice extent but bromine levels increase. We consider several possible explanations and ultimately suggest that increased acidity, specifically nitric acid, of snow on sea ice stimulates the release of reactive Br from sea ice, resulting in increased transport and deposition on the Greenland ice sheet.


Introduction 29
Atmospheric chemistry in the polar regions is strongly modulated by physical, chemical, and biological 30 processes occurring in and around sea ice. These include sea salt aerosol generation, biogenic emissions 31 of sulphur-containing gases and halogenated organics, and the photochemical/heterogeneous reactions 32 leading to release of volatile, reactive bromine species. The resulting chemical signals influence the 33 chemistry of the aerosol deposited on polar ice sheets. For this reason ice core measurements of sea salt 34 ions, methanesulphonate (MSA), and bromine have been examined as potential tracers for sea ice extent 35 (Abram et al., 2013;Spolaor et al., 2013bSpolaor et al., , 2016Wolff et al., 2003). The interpretation of such tracers 36 is complicated by the fact that their source functions reflect changes in highly complex systems, and 37 signals are further modified by patterns of atmospheric transport and deposition. 38 MSA is produced by the atmospheric oxidation of DMS (( # ) ( ). DMS is produced throughout the 39 world's oceans as a breakdown product of the algal metabolite DMSP, (( # ) (

Sampling and analysis 90
The ice cores were sampled from 33x33 mm cross-section sticks using a continuous melter system 91 (McConnell et al., 2002). The silicon carbide melter plate provides three streams from concentric square 92 regions of the ice core sample: an innermost stream (with a cross sectional area of 144 mm 2 ), an 93 intermediate stream (340 mm 2 ) and an outer stream that was discarded along with any contaminants 94 obtained from handling of the ice core. The innermost melt stream was directed to two inductively 95 coupled plasma-mass spectrometers (ICP-MS, Thermo Element II high resolution with PFA-ST 96 concentric Teflon nebulizer (ESI)) run in parallel. All calibrations and runtime standards were run on 97 both instruments and several elements were also measured in duplicate (Na, Ce, Pb) to ensure tracking 98 between both ICP-MS. In addition, an internal standard of yttrium flowed through the entire analytical 99 system and was used to observe any change in system sensitivity. The instrument measuring bromine 100 was run at low resolution to get the highest sensitivity and there were no mass interferences observed at 101 the bromine isotope mass monitored (79 amu). The sample stream was acidified to 1% # to prevent 102 loss of less soluble species, degassed just prior to analysis to minimize mixing in the sample line and 103 sampled at a rate of 0.  Table S1. 182 Sea-salt transport onto the Greenland ice sheet occurs predominantly during winter. Historically the 183 winter-time sea-salt maximum was believed to be due to increased cyclonic activity over the open 184 oceans (Fischer and Wagenbach, 1996) though more contemporary studies show that blowing snow 185 from the surface of sea-ice may be a significant source (Rankin et al., 2002;Xu et al., 2013;Yang et al., 186 2008Yang et al., 186 , 2010. At Summit, a winter-time maximum is observed in the most abundant sea salts, Na and Cl 187 (Fig. 3). Bromine also shows a significant winter-time signal, however the annual maximum appears in 188 mid-summer -at concentrations ~70% above winter levels (Fig. 3a). Comparison with Br measured in 189 weekly surface snow samples collected from Summit (from 2007-2013; GEOSummit project) confirms 190 that this summer signal is real and not a result of post-depositional modification of seasonality of the 191 bromine signal (Fig. S1). The results from that study confirm that total Br concentrations peak in 192 summer on the ice sheet closely following the Br cycle observed in the Summit-2010 ice core. In 193 addition to the comparison with the Geosummmit data, in the ice cores studied here there are routinely 194 more than 10 measurements made within a yearly layer of snow giving confidence to the allocation of 195 a summer maximum in bromine at Summit. Analysis of the annual cycle of bromine in the Tunu ice 196 core also shows a summer maximum when averaged over the entire ice core time series but with 197 significantly larger error than observed at Summit. The timing of this peak suggests a predominant 198 summer source of bromine that dwarfs that from winter sea salt sources. 199 The shape of the annual bromine cycle does change slightly over the course of the Summit record (see 200 the maximum from a solely summer peak in the preindustrial era towards a broad summer-spring peak 202 by 1970 is observed (Fig. 3 lower plot). Comparison with the sea salt tracer, sodium, which does not 203 undergo the large temporal shift and broadening of its seasonal cycle shows that this change in bromine 204 seasonality is not linked to changes in production or transport of sea-salt aerosols or even dating 205 uncertainties in the ice core but perhaps the introduction of an additional, smaller bromine source in the 206 spring-time during the industrial era. 207 Both ice cores show a predominantly positive Br enrichment throughout the year (Fig S2)  Comparison with the total sulfur record (Fig. 4) reveals that during the preindustrial period, MSA 245 contributes to ~12% and ~ 7% of the total sulfur signal at Summit and Tunu, respectively, compared 246 with < 2% at the height of industrial period (1970 C.E.) at both sites. 247 The low frequency, preindustrial trend in MSA concentrations seen in these ice core records closely 248 follows that of bromine; particularly distinct is the decrease in both MSA and bromine at both sites in 249 the early 1800s (Tables S1and S2). In the early 1900s, however, both sites show a divergence between 250 the MSA and Br records-as MSA begins to decline, Br concentrations increase. 251 A dramatic shift in the 'timing' of the annual MSA maximum in Summit-2010 ice core is illustrated in 252 Figs. 3c and S3. The signal shifts gradually and continuously along the length of the the entire Summit-253 2010 record from a spring to winter maximum (Fig. S3). This phenomenon has previously been 254 observed in several Antarctic ice cores and has been attributed to post-depositional migration within the 255 ice due to salt gradients (Mulvaney et al., 1992;Weller, 2004). 256

Acidic Species 257
In winter, with the collapse of the polar vortex, polluted air masses enter the Arctic region as the 258 phenomenon known as the Arctic haze (Barrie et al., 1981;Li and Barrie, 1993). Total snow acidity was stable at both sites from 1750 through to ~1900 except for sporadic, short-lived 265 spikes due to volcanic eruptions. The average preindustrial acidity was the same at both sites (~1.8 µM). 266 Both records also show two distinct maxima in acidity centred on 1920 and 1970 ( Fig. 4) with Tunu 267 displaying higher acidity than Summit over the entire industrial period. Overlaid with the acidity is the 268 total sulphur (S) record for both ice cores. The high correlation between the acidity and S records 269 illustrates that the sulphur species are the dominant natural and anthropogenic acidic species in the ice 270 cores. The trend in acidity closely follows the global ( emissions with maxima from coal (~1920) 271 and fossil fuel combustion (~1970), respectively (Smith et al., 2011). After 1970 the records of acidity and S deviate. This deviation can be attributed to the presence of nitric acid that remains at a relatively 273 high concentration in the late 20 th century whilst sulphur species reduce in concentration (Fig. 4). The results for Tunu indicate that air masses arrive primarily from the west coast of Greenland, passing 294 over the Baffin Bay area, but there is also significant contribution from both the SE and NE (in May) 295 coastal areas (Fig. 6b, S4b). Of these two secondary areas it is likely that aerosols transported from the 296 NE would have a greater influence on the ice core concentrations due to proximity to the ice core site. 297 Aerosol deposited at Tunu therefore represents a mixture of source regions, but are likely dominated by 298 the NW Greenland, Baffin Bay coastal region. 299

MSA -Sea Ice correlations 300
Locations which showed a sea ice concentration (SIC) variability greater than 10% (the average 301 estimated range of uncertainty in the satellite measurements) and have a significant correlation to MSA 302 (t-test, p<0.05) are displayed in Fig. 6 Table S1). This sea ice decline is coincident with the sustained increase 372 in greenhouse gases which has been identified as the major climate forcing and driver of increased 373 global temperatures during the 20 th century (Mann et al., 1998) and follows the same general trend in 374 Arctic wide sea ice extent observed by Kinnard (2008). 375 Bromine has also been suggested as a possible proxy for sea ice conditions, however the timing of the 376 largest bromine aerosol flux, in summer, does not coincide with the largest growth or extent of new sea 377 ice. Sea ice begins to increase only at the end of summer as the fractures in the ice cover are re-laminated 378 and the ice edge begins to advance southward (see Fig. 3f). 379 So what is the summer-time source of bromine? What is the cause of the increase in spring-time bromine 380 explosion events in the industrial era? (see Fig. 3

Leaded Gasoline 398
The largest historical, anthropogenic source of bromine is thought to be the combustion of leaded 399 gasoline. Large quantities of 1,2, diethyl bromide (DEB) were added to leaded fuel as a scavenger for 400 Pb preventing lead oxide deposition by converting it to volatile lead bromide salts as well as   by acidic sulphate aerosols, for example, is estimated at 80% compared to 30% for sea salt aerosols 482 (Parrella et al., 2012). Interestingly, Abbatt (1995) demonstrated that is more than 100 times more 483 soluble in super-cooled sulphuric acid solutions than . This may explain the cause of bromine 484 enrichment in the aerosol measured in the ice cores relative to the more abundant chlorine (Fig. S2). 485 The results of both the laboratory and field studies suggest that increasing snow/ice acidity in the Arctic 486 will likely enhance spring-time bromine explosion events above the sea ice whilst the increase in 487 solubility allows the termination products of the explosion to be transported away from the sites on the  Figure 9 illustrates that of the two dominant acidic species preserved in the ice, # (represented by 492 nitrate) shows the highest correlation to total bromine over sub-decadal time scales at both ice core sites. 493 Records were detrended with an 11 year running average before comparison to isolate the high 494 frequency components of each record. The bromine -sulfuric acid (represented by sulfate) correlation 495 is not significant. This is primarily because there is no bromine response to the dominant volcanic 496 sulphate spikes throughout the record. The large spikes in sulfate concentrations did not cause a 497 depletion of bromine in the snowpack (Figure 9). This result might be expected if the increased acidity 498 caused more bromine to volatize. These results suggest that # is the most influential of the MBL 499 acidic species in the processing and transport of Br on aerosols in the MBL. 500

NOx and links to bromine 501
The snow and atmospheric chemistries of bromine and nitrate ( its high solubility. The heterogeneous hydrolysis of # to again release bromine species back into 508 the gas-phase has also been observed (Parrella et al., 2012) and can occur both during sunlight hours as 509 well as in the dark (Sander et al., 1999).
s are intertwined with Br as it cycles between the gas and 510 condensed phases. 511 The seasonality of the # * signal preserved in the ice cores is coherent with Br, showing a summer-512 time maximum (Fig. 3a,d). The slight shift in timing of the industrial nitrate seasonal maximum towards 513 spring is replicated in the seasonal bromine signal preserved in the ice (Figure 3). The high correlation 514 between the preindustrial (1750-1850) # * and Br records (Fig. 9) supports this observation of co-515 transport and sink of Br and # * into the snow pack, though the natural sources of each are distinctly 516 different. 517 In the industrial era the low frequency temporal profile of the bromine (Fig. 2) and nitrate records ( gasoline (Sect. 4.1.2). As discussed above, we observe that leaded fuel pollution reaching the Arctic 524 began to decline after 1970 in-line with reduced global consumption, but the amount of bromine in-525 excess of natural sources (exBr, Fig. 8) continued to increase -following the trends in s pollution 526 (Fig. 4). The continued increase in s despite the decline in leaded fuel combustion is attributed 527 primarily to biomass burning, soil emissions and unleaded fossil fuel combustion (Lamarque et al., 528 2013). As the leaded fuel source of bromine began to decline, organic bromine pollutants continued to 529 increase, as was discussed in Sect. 4.1.4. This can only account for a small fraction of the observed Br. 530 The continued correlation between nitrate and exBr despite the decoupling of nitrate and bromine 531 anthropogenic sources after 1970, suggests that nitrate pollution is likely influencing the processing of 532 local, natural sources of bromine in the polar MBL, in effect increasing the mobility of the bromine and 533 thus its flux onto the ice sheet. 534

Consequences of nitrate driven increased bromine mobility in the Arctic 535
Plumes of BrO emitted from sea ice regions have been linked to mercury deposition events which lead 536 to an increase in the bioavailability of toxic mercury species in polar waters (Parrella et al., 2012). 537 Increased spring-time mobilization of bromine from the sea ice induced by anthropogenic nitrate could 538 therefore increase the frequency and duration of these events and thus the mercury toxicity of the oceans. 539 Increased atmospheric bromine concentrations would also increase the frequency of ozone depletion 540 events (Simpson et al., 2007) thereby altering the oxidative chemistry of the polar MBL. 541 Whilst several studies have begun to explore bromine records from ice cores as a proxy for past sea ice 542 conditions, the results of this study demonstrate that in an era of massive increases in atmospheric acidity 543 the natural relationship between bromine and sea ice conditions can become distorted, precluding it 544 from being an effective modern-day Arctic sea ice proxy. 545 546

Conclusion 547
In this study we have shown that high resolution MSA measurements preserved in ice cores can be used 548 as a proxy for sea ice conditions (specifically the size of the marginal sea ice zone) along specific 549 sections of the Greenland coast. The MSA records show that sea ice began to decline at the end of the 550 LIA and again, more dramatically during the Industrial period. Also, unsurprisingly, the changes in sea 551 ice conditions in the northern sites have been less dramatic than along the southern coastline.    Table S1. 825