Baseline Acidity of Percipitation at the South Pole During the Last Two Millennia J H

University of Nebraska - Lincoln DigitalCommons@University of Nebraska - Lincoln

US Army Research U.S. Department of Defense

1987 Baseline Acidity of Percipitation at the South Pole during the Last Two Millennia J H. Cragin U.S. Army Cold Regions Research and Engineering Laboratory

M B. Giovinetto University of Calgary

A J. Gow U.S. Army Cold Regions Research and Engineering Laboratory

Follow this and additional works at: http://digitalcommons.unl.edu/usarmyresearch Part of the Earth Sciences Commons, and the Environmental Sciences Commons

Cragin, J H.; Giovinetto, M B.; and Gow, A J., "Baseline Acidity of Percipitation at the South Pole during the Last Two Millennia" (1987). US Army Research. 308. http://digitalcommons.unl.edu/usarmyresearch/308

This Article is brought to you for free and open access by the U.S. Department of Defense at DigitalCommons@University of Nebraska - Lincoln. It has been accepted for inclusion in US Army Research by an authorized administrator of DigitalCommons@University of Nebraska - Lincoln. GEOPHYSICALRESEARCH LETTERS, VOL. 14, NO. 8, PAGES789-792, AUGUST1987

BASELINE ACIDITY OF PRECIPITATION AT THE SOUTH POLE DURING THE LAST TWO MILLENNIA

J.H. Cragin

U.S. Army Cold Regions Research and Engineering Laboratory

M.B. Giovinetto

University of Calgary, Alberta

A.J. Gow

U.S. Army Cold Regions Research and Engineering Laboratory

Abstract. Measurementsof meltwater pH from annual ward transport of pollutants from Eurasia and North layersof SouthPole firn and ice samplesranging in age America to the Arctic during winter [Rahn, 1981]. Because from 40 to 2000 yearsB.P. showthat precipitationat this of potential long-rangetransport, the possibilityalways remotesite has a highernatural aciditythan that expected existsthat present-dayprecipitation, however remote its from atmosphericequilibrium with CO2. The averagepH location, will contain anthropogenic H2SO4 or HNOs. of deaerated(CO2-free) samples was 5.64 + 0.08, whileair- One way to eliminatethe anthropogeniccomponent is equilibratedsamples averaged 5.37 + 0.08, a pH that is to usesamples of polar snowand ice whichwere deposited about a factor of two more acidic than the expected back- before the industrial revolution and are now preserved groundpH of 5.65. The observed"excess" acidity can be chronologicallyin permanentice sheets.Impurities are in- accountedfor by SO1-and NO7 levelsin the samplesorig- corporatedinto the snowby rainout,washout and dry fall- inating from non-anthropogenicH2SO4 and HNO3. Be- out. Polar snow and ice sampleshave been usedextensive- causeof the presenceof thesenaturally occurring acids in ly to provide natural backgroundconcentrations and en- SouthPole precipitation,a pH of 5.4 is considereda more richmentsof many pollutants, especiallymetals [Cragin et representativebaseline reference pH for acidprecipitation al., 1975; Herron et al., 1977; Boutron and Lorius, 1979; studies. Weiss et al., 1979; Boutron, 1980; Boutron, 1982]. Al- though ice core acidity has been estimatedindirectly by Introduction surface conductivity methods [Hammer 1980; Hammer et al. 1980],the only measurementsof pH on liquid samples Most studiesof acid precipitationhave used relatively reportedfor pre-1800ice samplesare thoseof Koernerand recent rain or snow samplesfrom either populated areas Fisher [1982], who found a mean Holocene (200-5500 [Cogbilland Likens, 1974; Galloway et al., 1976;Wolff et yearsB.P.) pH of 5.48 in coresfrom the AgassizIce Cap, al., 1977;Brezonik et al., 1980;Lewis and Grant, 1980]or Ellesmere Island, Canada. However, as they state, no at- remotelocations [Weiss et al., 19778Sequeira, 1981; Gal- temptwas made to eliminateor controlCO•-induced acid- lowayet al., 1982;.lickells et al., 1982;Koerner and Fish- ity, which could introducevariability in the background er, 1982;Pszenny et al., 1982;Keene et al., 1983].While pH results[Koerner and Fisher1982]. Titrations of acidity remoteprecipitation is generallyless acidic than that from in ancient(30,000 yearsB.P.) ice core samplesfrom Dome industrializedareas, even samplesfrom remote locations C, Antarctica (74ø39'S, 124ø10'E), gave an average H* are often more acidic than the pH 5.65 expectedfrom dis- concentrationof 2.44/zEq/L [Delmaset al., 1980], which solutionof atmosphericCO2. (A derivation of the effect is equivalentto a pH of 5.61. In similartitrations of seven of CO2 on the acidity of pure, i.e., distilled, water can be deeper(800 to 900 m) samplesfrom the Dome C ice core, foundin Craginet al. [ 1984].) A major concernis whether Delmas et al. [1980] obtained an averageacidity of 1.07 this excessacidity is of natural or anthropogenicorigin. + 0.52 tzEqH+/L, correspondingto a pH of 5.97. These Acid rain on Bermuda has been ascribed to long-range at- older samplescorrespond to Wisconsinage ice deposited mospherictransport of sulfatesand nitratesfrom North when the acid-neutralizingterrestrial dust input to ice America [.lickellset al., 1982]. Galloway et al. [1982] re- sheetswas considerablyhigher than during the last 10,000 centlymeasured the pH of precipitationfrom five remote years[Cragin et al., 1977; Briat et al., 1982]. Here we pre- areas of the world and attributed much of the acidity at sentvalues of pH from both air-saturatedand CO,-free theselocations to long-rangetransport. The well-known South Pole ice core samplesspanning the last 2000 years. arctic haze over Alaska is believed to be caused by north- Experimental

Copyright1987 by the AmericanGeophysical Union. A 200-m-long, 10-cm-diameterice core was recovered from the Amundsen-Scott South Pole Station during the Paper Number 7L8013. 1981-82 austral summerusing an electromechanicaldrill 0094-8276/87/007L- 801353.00 [Rand,1976; Kuivinen et al., 1982].The corewas "dated"

789 790 Craginet al.: BaselineAcidity of Precipitationat the SouthPole

stratigraphicallyalong its entirelength [Kuivinen et al., scribed elsewhere [Langway et al., 1974; Cragin et al., 1982] by continuouscounting of coarse-graineddepth- 1975; Herron et al., 1977]. Immediately after cleaning, hoarlayers that are knownto occuron an annuallyrepeat- these samplesof firn or ice were placed in individual wide- ing basis[Giovinetto, 1960; Gow, 1965].With this tech- mouth polyethylene (LPE) bottles and allowed to melt. nique,age errorsincrease with depthbecause of missed Several samples were initially melted under a nitrogen at- yearsattributable to hiatusesin accumulation.However, mosphere but pH values were similar to those obtained by the error is estimated not to exceed 100 years for the deep- nitrogen deaeration only, indicating that melting under ni- estand oldest(approximately 2000 years B.P.) portionof trogen was not necessary.After each melted sample had the core. reached room temperature (25 + 1 øC), it was deaerated by The core was transportedfrozen to the U.S. for subse- bubbling filtered (0.4 t•m) purified nitrogen gas through it quentanalysis at the U.S. Army ColdRegions Research for 15 minutes to remove CO2 and eliminate pH variability and EngineeringLaboratory (CRREL). For the chemical causedby differing amounts of H2CO3. Additional deaer- measurementsreported here, we used 10- to 20-cm-long ation (Table 1, column four) produced no significant pH half-coresegments, each representing exactly one year's change, indicating that 15 minutes was sufficient. The accumulation to eliminate chemical variability due to sea- sample pH was then measured under a nitrogen atmos- sonal concentration differences [Langway et al., 1975]. phere in a specially constructed Plexiglas chamber. Sam- The firstthree samples, from depthsof 7.72-7.92m, 14.63- ples were then allowed to equilibrate with air overnight in 14.82 m and 20.00-20.15 m are of snow deposited since the dark for several days and measurements were again the onset of the Industrial Revolution. Snow layers made to determine the pH of air-saturated samples. depositedsince 1940 (above 7.7 m) werenot sampledbe- All pH measurementswere made with an Orion model causeof possiblelocal contamination from the South Pole 81-03 combination Ross electrode. This electrode differs Station,constructed and occupiedsince 1957. Older sam- from ordinarycombination pH electrodesin that the ref- pleswere selected at 20-m intervals. erencehalf cell potential is controlledby reversibleredox All subsequentlaboratory work, includingsample clean- equilibriarather than by the conventionalcalomel or Ag/ ing, melting,deaeration and analysis, was performed in a AgC1.Compared to a conventionalcombination pH elec- Class 100 clean room. Utensilsand labware (Teflon, poly- trode, the Ross electrode respondedmore rapidly and ethylene)used to cleanthe coreor containsample melt- showednegligible drift evenin low ionicstrength solutions waterwere cleaned by rinsingfirst with distilled-deionized of South Pole meltwater. The precisionand accuracyof water,then by rinsingwith electronicgrade acetone and thesemeasurements is estimatedto be + 0.05 pH unit. finallyby leachingfor 3 dayswith 18 megohmMilli-Q © deionized water. Surface contaminants were removed Results and Discussion from sectionsof firn core samplesby mechanicallyscrap- ing off the outer 0.5-1.0 cm with a stainlesssteel chisel. The values for pH and correspondingcalculated acidi- This processwas conductedwithin a Class100 laminar- ties presentedin Table 1 are for 12 annual incrementsof flow clean air station in a coldroom using extreme care to South Pole core coveringapproximately the last two mil- allow cleaned areas of the core to contact only cleaned lennia. Important features of thesedata include the uni- Teflon (FEP) or polyethylene(LPE). Deepercore samples formity of pH valuesand the absenceof any appreciable (depthrange 120-200 m) consistingof impermeableglacial trend in pH over the core profile. The lack of an observed ice were cleaned in a Class 100 clean room by repeated increasein acidity (decreasein pH) is probably due to the washingswith 18 megohmMilli-Q © deionizedwater as de- age of the samples;the most recentsample fell as snow40

Table 1. Valuesfor pH and correspondingacidities for firn and icecore samples from the South Pole, Antarctica

Sample Agein years N2 atmosphere Air depth before1982 pH after pH after [H*] [H*] (m) (est.error + 5ø70) 15min. 30 min. (/•M/L) pH (/•M/L)

7.72- 7.92 40 5.62 -- 2.40 5.37 4.27 14.63- 14.82 90 5.61 -- 2.45 5.34 4.57 20.00- 20.15 130 5.57 • 2.69 5.42 3.80 39.72- 39.84 270 5.72 • 1.91 5.37 4.27 59.00- 59.13 440 5.73 • 1.86 5.48 3.31 79.05- 79.15 640 5.69 • 2.04 5.42 3.80 101.26-101.39 860 5.51 -- 3.09 5.33 4.68 120.54-120.65 1070 5.68 • 2.09 5.38 4.17 141.00-141.15 1300 5.83 5.79 1.62 5.40 3.98 160.00-160.12 1530 5.90 • 1.26 5.38 4.17 179.12-179.27 1750 5.47 • 3.39 5.24 5.75 200.39- 200.53 2000 5.64 5.62 2.40 5.32 4.79 Mean and std. deviation* 5.64 _+0.08 2.27 _+0.60 5.37 _+0.08 4.30 _+0.62

* Standarddeviations represent the variabilitybetween sample concentrations rather than analytical precision. Cragin et al.: BaselineAcidity of Precipitation at the South Pole

years ago, before "tall stacks" of the 1950s exacerbated transport may dominate for our samples sincewe specific- the SO2 transport problem. Furthermore, since acid pol- ally avoided selectingice deposited during or shortly after lutants are generated mainly in the Northern Hemisphere, known volcanic events. Furthermore, aerosol concentra- they may not reach the Antarctic Ice Sheet in measurable tion measurements of Tomoyukuto et al. [1986] suggest amounts becauseof poor interhemispheric mixing and be- that stratosphere to troposphere particle transport over cause of washout and dry deposition during extended Antarctica is less important than horizontal tropospheric transport to Antarctica. This is supported by the absence transport of particles to the Antarctic. of an increase of sulfate ion, a major component of acid On polar ice sheetssnow is transformed into ice by com- precipitation, in cores covering the last 100 years from paction of the permeable firn that composesthe upper 115 Dome C, Antarctica [Delmas et al., 1980; Delmas and m at the South Pole. However, there is no evidence to indi- Boutron, 1980]. These South Pole pH values thus provide cate any significant change in the acidity of the snow a baseline or reference pH against which present-day pre- caused by post-depositional diffusion of air through the cipitation can be compared. permeable firn. Observationsin support of this include the Another important aspectof the data is the actual mag- very sharp nature of changesin the chemical composition nitude of the pH values themselves. For the air-equili- [Boutron, 1982] of firn from the South Pole and Dome C, brated samples, the average pH is 5.37 + 0.08--about Antarctica. Broader peaks would have been expected if twice as acidic as the 5.65 pH expected for "pure" rain- post-depositional diffusion of air through the firn was water in equilibrium with atmospheric CO2. The average contributing significantly to the measured chemical con- pH of deaerated samplesis 5.64 + 0.08. The difference of centrations. 0.27 pH units between aerated and deaerated samples is Alkaline minerals contained in airborne particulate mat- about what would be expected from dissolution of atmos- ter can neutralize acidity in precipitation, resulting in pheric CO2; air-saturated water contains 2.2 x 10-6 M H*/L lower background acidity (higher pH). Greater dust con- from H•CO3 dissociation, which would lower a 5.65 pH tent may be the reason for higher pH values found in an- sample by 0.28 pH units. Converting the above aerated cient ice samplesfrom Baffin Island by Koerner and and deaerated pH values to H* concentrations (Table 1, Fisher [1982]. But the concentration of terrestrial dust at columns five and seven) shows that half of the acidity is the SouthPole is verylow (lessthan 0.25 •Eq Ca + Mg/L) caused by CO• dissolution and half by other naturally oc- [Boutron,1982] and most of the Ca andMg presentprob- curring acids. Preliminary ion chromatographic measure- ably originates from sea salt rather than from acid- ments of SOi- and NO] concentrations in eight samples neutralizingminerals. Although no singlebackground pH gave concentrations of 1.0-2.1 /•Eq SOi-/L and 1.5-2.2 may be applicableto all locationsof the globebecause of t•Eq NO•/L. Based on Na* concentrations (0.57-1.4/•Eq/ varyingalkaline dust inputs [Gallowayet al., 1982] the L) measured in four samples, the sea salt contribution to acidityof SouthPole samplesis a goodrepresentation of SOF is calculated not to exceed 10ø70of the total SOF background levels without substantial influence from presentin the samples.The averagesum of SOi- and NO; wind-transported dust. Samples from less remote loca- concentrations, 3.3 •Eq/L, is more than enough to bal- tions on the globe, with local or regionaldust sources, ance the average2.3 •Eq/L of H* (pH 5.64) in the deaerat- would be expectedto have higher pH values for similar ed samples.Thus, on a molar equivalent basis, the pH of acid inputs. deaeratedsamples can easily be accountedfor by the SO4•- Conclusion and NO• concentrations if these ions originate from H•SO4 and HNO3. Some acidity may also be contributed Our results, showingSouth Pole ice depositsto be a fac- by weak organic acids (acetic, formic), but we have no ex- tor of two or more acidic than the presumed pH of 5.65, perimental evidenceto support this. confirm the postulation of Charlson and Rodhe [1982] Electrolytic conductivities of the deaerated samples that the natural acidity of dust-freeprecipitation is indeed ranged from 0.10 to 0.17 mS/m and correlated well (r = greater than that expectedfrom carbonic acid equilibrium 0.87, p = 0.01 for 6 d.f.) with values calculated from the alone. Thus, in assessinganthropogenic contributions to observed pH's and ion concentrations. Most of the back- present-dayprecipitation, the background reference pH ground H•SO4 presentin the ancientprecipitation samples should take into account the contribution of H* from not reachesthe South Pole through advection as well as subsi- only H2CO3, but other naturally occurring acids as well. dence from the upper troposphereand possiblythe lower Acknowledgements. The National ScienceFoundation stratosphere,which facilitates a direct contribution from provided logisticalsupport. The authorsthank the Polar the "Junge layer," a good sourceof H* that is replenished Ice Coring Office of the Universityof Nebraskaand Dr. by eruptive and nonviolent volcanic activity [Berresheim GeraldHoldsworth of EnvironmentCanada for procuring and Jaeshke, 1983]. However, Legrand and Delmas [1986] the 200-m South Pole core. The authors also thank Eric found no evidence for a stratosphericinput of nitrate in Peakeof the Universityof Calgaryfor preliminaryanaly- Antarctic snow and concluded that this ion originates sesof SOi- and NOj, and Dr. JosephL. Oliphantand from lightning at midlatitudes. It must then presumablybe Daniel C. Leggettof CRREL for reviewingthe manu- transported tropospherically to the Antarctic. Tropo- script. spherictransport paths are also believed to exist for con- References tinental aerosols [Polian et al., 1986] and sea salt [Bod- haine et al., 1986]. Sulfate, most of which is non-marine at Berresheim, H. and W. Jaeshke, The contribution of vol- the South Pole, can be transported through both tropo- canoes to the global atmospheric sulfur budget. J. Geo- sphericand stratosphericpathways, although tropospheric phys. Res., 88, 3732-3740, 1983. 792 Craginet al.' BaselineAcidity of Precipitationat theSouth Pole

Bodhaine, B.A., J.J. Deluisi, J.M. Harris, P. Houmere land ice sheet evidence of post-glacial volcanism and its and S. Bauman, Aerosol measurements at the South climatic impact. Nature, 288, 230-235, 1980. Pole. Tellus, 38B, 223-235, 1986. Herron, M.M., C.C. Langway, Jr., H.V. Weiss and J.H. Boutron, C., Respectiveinfluence of global pollution and Cragin. Atmospheric trace metals and sulfate in the volcanic eruptions on past variations of the trace metals Greenland Ice Sheet. Geochim. et Cosmochim. Acta, content of Antarctic snows since 1880's. J. Geophys. 41, 915-920, 1977. Res., 85, 7426-7432, 1980. Ito, T., Y. Morita and Y. Iwasaka. Balloon observations Boutron, C., Atmospheric trace metals in snow layers de- of aerosols in the Antarctic troposphere and strato- posited at the South Pole from 1928 to 1977. Atmos. sphere. Tellus, 38B, 214-222, 1986. Env., 16, 2451-2459, 1982. Jickells, T., A. Knap, T. Church, J. Galloway and J. Mil- Boutron, C. and C. Lorius, Trace metals in Antarctic ler. Acid rain on Bermuda. Nature, 297, 55-57, 1982. snow since 1914. Nature, 277, 551-554, 1979 Keene, W.C., J.N. Galloway and J.D. Holden, Jr. Meas- Brezonik, P.L., E.S. Edgerton and C.D. Hendry. Acid urement of weak organic acidity in precipitation from precipitation and sulfate deposition in Florida. Science, remote areas of the world. J. Geophys. Res., 88, 5122- 208, 1027-1029, 1980. 5130, 1983. Briat, M., A. Royer, J.R. Petit and C. Lorius. Late-glacial Koerner, R.M. and D. Fisher. Acid snow in the Canadian input of eolian continental dust in the Dome C ice core: high Arctic. Nature, 295, 137-140, 1982. Additional evidencefrom individual microparticleanal- Kuivinen, K.C., B.R. Koci, G.W. Holdsworth and A.J. ysis. Ann. Glaciol., 3, 27-3-1, 1982. Gow. South Pole ice core drilling, 1981-82. Ant. J. Charlson,C. and H. Rodhe.Factors controlling the acidi- U.S., 17, 89-91, 1982. ty of natural rainwater. Nature, 295, 683-685, 1982. Langway, C.C., Jr., M.M. Herron and J.H. Cragin. Cogbill, C.G. and G.E. Likens, Acid precipitationin the Chemical profile of the Ross Ice Shelf at Little America northeastern United States. Water Res. Res., 10(6•, V, Antarctica. J. Glaciol., 13, 431-435, 1974. 1133-1137, 1974. Langway, C.C., Jr., J.H. Cragin, G.A. Klouda and M.M. Cragin, J.H., M.M. Herron and C.C. Langway, Jr. Herron. Seasonal variations of chemical constituents in Chemistry of 700 years of precipitation at DYE-3, annual layers of Greenland deep ice deposits.US Army Greenland.US Army Cold RegionsResearch and Engi- Cold RegionsResearch and Engineering Laboratory Re- neeringLaboratory ResearchReport 341, 18 p., 1975. search Report 347, 5 p., 1975. Cragin, J.H., M.M. Herron, C.C. Langway, Jr. and G.A. Legrand, M.R. and R.J. Delmas. Relative contributions Klouda. Interhemisphericcomparison of changesin the of tropospheric and stratospheric sources to nitrate in composition of atmospheric precipitation during the Antarctic snow. Tellus, 38B, 236-249, 1986. late CenozoicEra. Proc. Polar OceansConf., Mont- Lewis, W.M., Jr. and M.C. Grant. Acid precipitation in real, Canada. Arctic Institute of North America, 617- the western United States. Science, 207, 176-177, 1980. 631, 1977. Polain,G., G. Lambert,B. Andouinand A. Jegou.Long- Cragin, J.H., M.B. Giovinetto and A.J. Gow. Baseline range transport of continental radon in subantarcticand acidity of ancientprecipitation from the South Pole. US antarctic areas. Tellus, 38B, 178-189, 1986. Army Cold Regions Researchand EngineeringLabora- Pszenny,A.P., F. MacIntire and R.A. Duce. Sea-saltand tory Report 84-15, 7 p., 1984. the acidity of marine rain on the windward coast of Delmas, R.J., A. Aristarian and M. Legrand. Acidity of Samoa. Geophys. Res. Let., 9, 751-754, 1982. Antarctic snow: A natural referencelevel for acid raias, Rahn, K.A. Relative importancesof North America and Proc. Int. Conf. Ecol. Impact of Acid Precip., March, Eurasia as sources of Arctic aerosol. Atmos. Env., 15, Sandefjord, Norway (D. Drables and A. Tollan, Ed.). 1447-1455, 1981. Norway: SNSF Project, 104-105, 1980. Rand, J.H. The USA CRREL shallow drill. In Ice-Core Delmas, R. and C. Boutron. Are the past variations in Drilling (J.F. Splettstoesser,Ed.). Lincoln: University stratospheric sulphate burden recorded in central Ant- of Nebraska Press, 133-137, 1976. arctic snow and ice layers? J. Geophys. Res., 85, Sequeira,R. Acid rain' Somepreliminary results from glo- 5645-5649, 1980. bal data analysis.Geophys. Res. Let., 8, 147-150, 1981. Galloway, J.N., G.E. Likens and E.S. Edgerton. Acid Weiss, R.E., A.P. Waggoner, R.J. Charlson and N.C. precipitationin the northeasternUnited States:pH and Ahlquist. Sulfate aerosol:Its geographicalextent in the acidity. Science, 194, 722-724, 1976. midwestern and southern United States. Science, 195, Galloway, J.N., G.E. Likens. W.C. Keene and J.M. Mil- 979-980, 1977. ler. The compositionof precipitationin remote areasof Weiss, H.V., M.M. Herron and C.C. Langway, Jr. Natu- the world. J. Geophys.Res., 87, 8771-8786, 1982. ral enrichment of elements in snow. Nature, 274, 352- Giovinetto, M.B. Glaciologyreport for 1958, South Pole 353, 1979. Station. Columbus, Ohio State University Research Wolff, G.T., P.J. Lioy, H. Golub and J.S. Hawkins. Acid Foundation Report 825-2-Part 4, 1960. precipitationin the New York metropolitanarea: Its re- Gow, A.J. On the accumulation and stratification of snow lationship to meteorologicalfactors. Env. Sci. Tech., at the South Pole. J. Glaciol., 5, 467-477, 1965. 13, 209-212, 1977. Hammer, C.U. Acidity of polar ice coresin relation to ab- solute dating, past volcanism, and radio-echoes.J. Gla- (Received April 22, 1987; ciol., 25, 359-372, 1980. revised May 15, 1987; Hammer, C.U., H.B. Clausenand W. Dansgaard.Green- acceptedJune 23, 1987.)