Regional Distribution of Monsoon and Desert Dust Signals Recorded in Asian Glaciers Cameron P

The University of Maine DigitalCommons@UMaine

Earth Science Faculty Scholarship Earth Sciences

7-23-1993 Regional Distribution of Monsoon and Desert Dust Signals Recorded in Asian Glaciers Cameron P. Wake

Paul Andrew Mayewski University of Maine, [email protected]

Xie Zichu

Wang Ping

Li Zhongquin

Follow this and additional works at: https://digitalcommons.library.umaine.edu/ers_facpub Part of the Climate Commons, Geochemistry Commons, Glaciology Commons, and the Hydrology Commons

Repository Citation Wake, Cameron P.; Mayewski, Paul Andrew; Zichu, Xie; Ping, Wang; and Zhongquin, Li, "Regional Distribution of Monsoon and Desert Dust Signals Recorded in Asian Glaciers" (1993). Earth Science Faculty Scholarship. 190. https://digitalcommons.library.umaine.edu/ers_facpub/190

This Article is brought to you for free and open access by DigitalCommons@UMaine. It has been accepted for inclusion in Earth Science Faculty Scholarship by an authorized administrator of DigitalCommons@UMaine. For more information, please contact [email protected]. GEOPHYSICAL RESEARCH LETTERS, VOL. 20, NO. 14, PAGES 1411-1414,JULY 23, 1993

REGIONAL DISTRIBUTION OF MONSOON AND DESERT DUST SIGNALS RECORDED IN ASIAN GLACIERS

CameronP. Wake andPaul A. Mayewski

GlacierResearch Group, University of New Hampshire

Xie Zichu,Wang Ping and Li Zhongqin

LanzhouInstitute of Glaciologyand Geocryology, China

Abstract. Short-term(6 monthsto 17 years)glaciochemical RegionalSetting recordshave been collectedfrom glacierbasins throughout the mountainsof centralAsia. The spatialdistribution of The glaciochemicalrecords discussed here cover a wide snowchemistry in centralAsia is controlledpredominantly geographicarea (Figure 1), andcan be separatedinto four by the influxof dustfrom the aridand semi-arid regions in distinct regionson the basisof landscape,as defined by centralAsia. The glaciochemicaldata suggests that glaciers Alekseyevet al., (1988), and/orclimatic regime as reviewed which are removed from large sourceareas of mineral by Ramage (1971), Barry and Chorley (1982), and aerosol,such as thosein the Himalaya,the Karakoram,and Hastenrath(1985): the southeasternTibetan Plateau,are the onesmost likely to (i) NgozumpaGlacier (site 1) lies on the southernslopes containlonger-term glaciochemical records which detail of the easternHimalaya. This region is characterizedby annual to decadal variation in the strengthof the Asian mixed forest and small scale agriculturebelow 4500 m. monsoonand long-range transport of Asiandust. Precipitationis derivedfrom monsoonalair massesduring the summerand from westerlydepressions during the winter. Introduction (ii) SentikGlacier (site2) andHispar Glacier (site 3) lie in the western Himalaya and Karakoram, respectively. Both marineaerosol transported inland with the Asian Moisture derived from westerly air massesthroughout the monsoon and dust derived from the vast arid and semi-arid year results is high snow accumulationin this region regionsof centralAsia play majorroles in geochemical (Mayewskiet al., 1984;Wake, 1989).The westernHimalaya cyclesin Asia(e.g., Sequeira and Kelkar, 1978; Galloway et and, to a lesserextent the Karakoram,are also influencedby al., 1987; Prosperoet al., 1985). Although regional precipitationderived from summertime monsoon circulation. precipitationchemistry investigations have been performed in In addition, add and semi-addregions, including the Thar India andeastern China, little is knownconcerning the spatial Desert(Figure 1), lie to the southwestof the Karakoramand and temporalvariability of precipitationchemistry on the the westernHimalaya. Tibetanplateau and bordering mountain areas (Harte, 1983; (iii) XixabangmaPeak (site 4), QiangYong Glacier (site Wakeet al., 1990).Here we reporton a suiteof short-term(6 5), and Mt. Geladaindong(site 6) fall within the grassland monthsto 17 years)glaciochemical data sets which describe steppeswhich dominatethe southeasternTibetan Plateau. the spatialvariation of snowchemistry throughout central This regionreceives most of its precipitationfrom summer Asia. This data baseis usedto quantifythe influenceof plateaumonsoon circulation (Murakami, 1976). marine aerosol and Asian dust on the chemical content of (iv) The northernand westernregions of the Tibetan snowin the mountainsof centralAsia, and identify glaciers Plateau are characterizedby semi-arid and arid regions. whosechemical records are sensitiveto the annualvariability MeikuangGlacier (site 7) andthe Dunde Ice Cap(site 11) lie of theAsian monsoon and long-range transport of Asiandust. adjacentto the QaidamBasin, and MustaghAta (site 8) ChongceIce Cap (site 9) and GlacierNo. 1 (site 10) lie Samplingand Analysis adjacentto the Taklamakandesert. Surface material in the Taklamakanand Thar desertsis typicallycalcareous (Dregne, Over the past severalyears we have collectedhigh 1968; McKee, 1979). Numerouslarge depressionsin the resolutionglaciochemical records from Sentik Glacier (site 2 TaklamakanDesert and QaidamBasin contain extensive salt in Figure1)(Lyons and Mayewski, 1983; Mayewski et al., depositsrich in sodiumand chloride (Dregne, 1968; Chen 1983; 1984), Biafo Glacier(site 3)(Wakeet al., 1989), and and Bowler, 1986). In addition,dust storms occur frequently Glacier No. 1 at the headwatersof the Urumqi River (site in the add and semi-aridregions of centralAsia duringthe 10)(Wakeet al., 1992).More recently,snow and ice samples spring(Middleton et al., 1986). havebeen collected from NgozumpaGlacier (site 1), Hispar Glacier (site 3), XixabangmaPeak (site 4), Qiang Yong Results Glacier(site 5), Mt. Geladaindong(site 6), MeikuangGlacier (site7), MustaghAta (8), andChongce Ice Cap (site9). The majorion contentof centralAsian snow displays a Extremecare was taken at all timesduring sample collection systematicregional variation (Table 1). Glaciersin the andhandling to assuresamples remained uncontaminated. northernand western Tibetan plateau (Figure 1, sites7 to 11) Snowsamples were analyzedfor anions(chloride, sulfate, exhibit geometricmean concentrationsof calciumgreater and nitrate)and cations(sodium, potassium, calcium, and than 24 geq kg-1, and of sodium,chloride, and sulfate magnesium)using a DionexIon Chromatograph model 2010. greaterthan 2.4 geq kg-1. Glaciersin the southeastern Detailedmethods appear elsewhere (Wake et al., 1992).We Tibetan Plateau (sites 4, 5, and 6) display calcium levels alsoinclude snow chemistry data from Dunde Ice Cap(site rangingfrom 1 to 8 geqkg -1, andsodium, chloride, and 11)(Thompsonet al., 1990). sulfatelevels ranging from 0.4 to 1.5geq kg -1. Glaciers in the westernHimalaya and Karakoram(sites 2 and 3) show geometricmean calcium concentrations of7.3 geq k.g -1 (site Copyright1993 by theAmerican Geophysical Union. 3 only)and sodium, chloride, and sulfate levels ranging from 0.5 to 2.0 geq kg-1. NgozumpaGlacier (site 1) on the Papernumber 93GL01682 southernslope of theeastern Himalaya shows concentrations 0094- 8534/93/93 GL-01682503.00 of calcium, sodium,chloride, and sulfate,less than 0.4 geq

1411 1412 Wakeet al.:Monsoon and Dust Signals in AsianGlaciers

40 ø 40 ø

30 ø 30 ø

80 ø 90 ø I00 o

I Glacier O Glaciochemical Sample Collection Sites

i:.'"•Sand, Desert 1 Ngozumpa Glacier- Eastern Himalaya 7 Meikuang Glacier- Eastern Kun Lun 2 Sentik Glacier- Ladakh Himalaya 8 Mustagh Ata - Pamirs • SaltFlat 3 Hispar Glacier- Karakoram 9 Chongce Ice Cap - Western Kun Lun 4 Xixabangma - So. Tibetan Plateau 10 Glacier No. 1 - Tien Shah 3 km Topographic 5 Qiang Yong Glacier - So. Tibetan Plateau 11 Dunde Ice Cap - Qilian Shah ------• Contour Line 6 Mt. Geladaindong - Tanggula Shah Fig. 1 Locationmap for glaciochemicalcollection sites in themountains of centralAsia. The heavydashed line representsme 3000 m contourline andoutlines the TibetanPlateau. Landscapeand topographicinformation were derivedfrom a map publishedby the RoyalGeographic Society (1987). kg-1. In contrastto thechemical species discussed above, -chlorideas a resultof the influx of desertdust. The highest nitrate concentrations show less variation in central Asian ion burdenswere measuredat Chongceand Dunde ice caps, snow, indicative of a separateand more distant source. which are close to (i.e., < 200 km) and downwind of the However, higher concentrationsof nitrateare still associated TaklamakanDesert and QaidamBasin, respectively. Sodium with higherlevels of calcium,sodium, chloride and sulfate. andchloride signals in the northeasternregions of theTibetan Measurementsof the annualaccumulation of majorions Plateau which could potentially relate to monsoon is only availablefor sevensites (Table 2). The annualflux of precipitation(i.e., peaksof =0.5 geqkg -1, seebelow), are ionsmeasured in centralAsian snow displays similar spatial effectively maskedby sodiumand chloridederived from the patternsto thoseapparent in the ion concentrationdata. arid and semi-aridregions in China. Mustagh Ata (site 8) in the westernTibetan Plateaushows The high levels of sulfatein snowfrom the northernand low annual snow accumulationrates but relatively high western Tibetan Plateau are associated with dust related annual fluxes of calcium, sodium, chloride, and sulfate. species(e.g., calcium, sodium,and chloride).While arid and Annualion fluxesin the southeasternTibetan Plateau (site 4, semi-addregions are potential source regions for sulfate(e.g. 5 and6) areintermediate between the high values at Mustagh Jungeand Werby, 1958;Barrie and Hales, 1984),temporal Ata andthe very low valuesat NgozumpaGlacier. Glaciers variations in sulfate in central Asian snow do not correlate in the Karakoramand western Himalaya (site 2 and3) show well with those in calcium, sodium and chloride. high annualsnow accumulation rates and high annual fluxes Anthropogenicsources of sulfateare recordedin snowfrom of calcium,sodium, chloride, sulfate, and nitrate. Ngozumpa northwestChina (Wake et al., 1992) andin snowthroughout Glacier (site 1) displaysintermediate value for annualsnow the former Soviet Union (Belikova et al., 1984). accumulation,but very low annualfluxes for all ions. Furthermore, Savoie et al. (1989) suggestthat su!fate associated with Sahara dust is derived not from the soil Discussion material, but from anthropogenicemissions in Europe and Africa. At presentwe are uncertainof therelative importance Snow and ice chemistryat any given site in centralAsia of desertdust and anthropogenicsources with respectto is closelyassociated with' proximity to arid and semi-arid sulfatedepostion in centralAsia. regions combined with atmospherictransport pathways. Lower concentrations of calcium, sodium, chloride and Glaciersin the northernand westernTibetan plateau, which sulfate in the southeasternTibetan Plateau(sites 4, 5, and 6) are surroundedby large desertbasins (sites 7 through 11), are indicativeof a decreasein the mineralaerosol input as a show elevated concentrations of calcium, sodium and resultof longertransport distance from largedesert basins in Wake et al.: Monsoonand Dust Signals in AsianGlaciers 1413

Table 1. Summaryof geometricmean values of majorion concentrationsin centralAsian snow

site mountain/elev periodof (geqkg-1) Fig. 1 glacier (m asl) record* n Ca++ Na+ C1- SO4= NO3- Na:CI@

SOUTH SLOPE- EASTERN HIMALAYA 1 Ngozumpa 5700 1989-90 30 0.40 0.31 0.33 0.24 0.11 0.98 + 0.25 WESTERN HIMALAYA & KARAKORAM 2 Sentik** 4908 1963-80 99 na 1.6 1.8 na 0.98 0.99 + 0.57 3 Hispar 5150 1985-88 76 7.3 0.52 0.89 1.5 2.0 0.65-+ 0.25 SOUTHEASTERN TIBETAN PLATEAU 4 Xixabangma 6140 1990-91 31 1.2 0.37 0.54 0.45 0.55 0.97 + 0.32 5 QiangTong 5850 1990-91 17 1.2 0.38 0.49 0.36 0.62 0.85 -+0.34 6 Geladaindong5950 1988-90 40 7.8 1.4 1.3 1.4 1.5 1.1 ñ0.26 NORTHERN & WESTERN TIBETAN PLATEAU 7 Meikuang 5520 Summr'91 12 24 6.8 7.5 4.5 2.1 0.91 + 0.06 8 MustaghAta 5910 1991-92 30 31 2.4 2.5 5.6 4.0 0.97 -+0.18 9 Chongceõ 6327 1980-87 77 64 18 17 16 6.4 1.1 ñ0.38 10 GlacNo. 1'* 3960 spring'89 27 47 8.0 5.5 12 3.1 2.8 +0.38 11 Dunde* 5325 1967-87 10 80 na 29 16 3.6 na

geometricmeans are used as the data from each site approaches a lognormal distribution. n is the numberof samplesanalyzed; here, na meansnot available. * Note thatthe periodof recorddiffers between sites. However,the 1-to-3fold variationfrom year-to-yearin the geometricmean ion concentrationsof snowat sitesfor which thereis morethan one year of record(i.e., Sentik, Hispar,Geladaindong, Mustagh Ata) is minimalcompared to the muchlarger variability observed on a spatial scale,suggesting that the data does provide a representativeregional perspective. ** SentikGlacier data from Mayewski et al., 1984;GlacNol datafrom Wake et al., 1992;Dunde Ice Capdata from Thompsonet al., 1990;the only publishedcalcium data from Dundeis for circa 1750. õ Thechemical record from a fun/icecore recovered from Chongce Ice Cap(Han et al., 1989)is discontinuousdue contaminationof thefire samplesduring transport. Only ion data from the ice sectionsof thecore are reported. @ TheNa:C1 ratio is given in geq ñ 1 standarddeviation. The ratio from sites 1 through9 approaches thatof seawater (0.86) andhalite (1.0). The highratio for site10 reflectsinputs of Na+ richcrustal material (Wake et al., 1992). western China. The vertical distribution of calcium, sodium, chemical influence of Asian dust. Sodium and chloride and chlorideat site 4, 5, and 6 displaystrong seasonality. concentrationsshow peaks in summerlaye.rs rising 0.5 geq Peaksis all threespecies occur in the spring/summerlayers kg-1 abovebaseline values (--0.3 geq kg-1), reflecting the of the snowpacksuggesting that dust raised during the spring influence of precipitation derived from monsoonalair dust storm period is transportedsouthward from the masses.The observedchemical patterns are consistentwith Taklamakan Desert and the Qaidam Basin by persistent local (Inoue, 1976; Yasunari, 1976)and synoptic(Luo and northwesterly surface winds (Luo and Yanai, 1983; Yanai, 1983; Murakami, 1987)meteorology. Murakami,1987). Higher concentrations and annual fluxes of calcium, sodium and chloride at Mt. Geladaindongare consistentwith thisinterpretation, as this glacier lies closer to Table 2. Annualaccumulation of snowand major ions the mineral aerosolsource regions to the north and west. in central Asia Summertimepeaks which are uniqueto the sodiumand chloride time-series, as well as summersnow accumulation, mountain/ snow (neqcm-2 yr-1) reflect the influx of marine air masses associated with plateaumonsoon circulation (Murakami, 1987). glacier n acc.* Ca++ Na+ C1- SO4= NO3- The high annualflux of snowand major ions at Hispar Glacier(site 3), andto a lesserextent at SentikGlacier (site SOUTH SLOPE- EASTERN HIMALAYA 2), suggeststhat much of the aerosol and moisture Ngozumpa 1 56 26 22 23 13 9 transportedwith the westerlyjet streamis removedas it ascendsthe southwest margin of theTibetan Plateau. Sodium WESTERN HIMALAYA & KARAKORAM andchloride show peaks in spring/summerlayers reflecting Sentik 17 60 na 152 146 na 70 inputsof marineaerosol from monsoonal air massesand/or Hispar 3 130 1580 273 146 273 390 inputsof evaporite-richdust transported from the Thar desert duringthe springtime dust storm period. SOUTH-EASTERN TIBETAN PLATEAU In contrastto sites2 through11, snowfrom Ngozumpa Xixabangma 1 68 169 34 43 45 45 Glacier (site 1) on the southern slopes of the eastern QiangYong 1 62 137 24 30 24 44 Himalayasdisplays very low concentrationsand annual Geladaindong2 30 794 78 64 62 74 fluxesof major ions. Ion concentrationsand annualfluxes arecomparable to thosemeasured in pre-1900AD snowfrom NORTHERN & WESTERN TIBETAN PLATEAU Summit Greenlandand in snowcollected 38 km upwind of MustaghAta 2 24 1400 115 108 182 93 the South Pole (Whitlow et al., 1992). Low calciumlevels suggestthat high elevation glaciers on the southern slopes of n is the numberof yearsof record;na meansnot available theeastern Himalaya are relatively free from the dominating * snowaccumulation isreported in g cm-2yr-1 1414 Wake et al.: Monsoonand Dust Signalsin AsianGlaciers

Conclusions KinematicAnalyses, Mon. Weath.Rev. 111, 922-944, 1983. The glaciochemicaldata set presented here confirms that Lyons,W. B. andP. A. Mayewski,Nitrate plus nitrite snowchemistry over the Tibetanplateau is dominatedby concentrationsin a Himalayanice core,Geophys. Res. desertdust derivedfrom the vast arid and semi-addregions Lett. 10, 1160-1163, 1983. of centralAsia, while snowfrom the southernslopes of the McKee, E. D., A Studyof Global SandSeas. Geological easternHimalaya is characterizedby very low ion burdens SurveyProfessional Paper 1052; UnitedStates Printing representativeof relatively clean free tropospheric air. Asian Office, Washington,1979. monsoonsignals are mostclearly recorded at glaciers,such Mayewski,P. A., W. B. LyonsW. B. andN. Ahmad, as the southernslopes in the easternHimalaya, which are Chemicalcomposition of a highaltitude fresh snowfall in well removed from the influence of Asian dust. Glaciers in the LadakhHimalayas, Geophys. Res. Lett. 10, 105-108, the westernHimalaya, the Karakoram,and the southeastern 1983 Tibetan Plateau record combined monsoonal and Asian dust Mayewski,P. A., W. B. Lyons,N. Ahmad,G. Smithand M. signals,while glaciers in thenorthern and western regions of Pourchet,Interpretation of the chemicaland physical theplateau contain records dominated by dustderived from time-series retrieved from Sentik Glacier, Ladakh surroundingarid and semi-aridregions. High elevation Himalaya,India, J. Glaciol.30, 66-76, 1984. glaciers in the Himalayas, the Karakoram, and the Middleton, N.J., A. S. Goudie and G. L. Wells, The southeasternTibetan plateau are thereforemost likely to frequencyand sourceareas of duststorms. in Aeolian containlonger-term glaciochemical records which reflect Geomorphology,edited by W. G. Nickling,237-259. regionalto hemisphericatmospheric chemistry and climate Allen & Unwin, Boston, 1985. signals,such as the annual to decadalvariation in thestrength Murakami, T., Effects of the Tibetan Plateau;in Monsoon of the Asianmonsoon and the long-rangetransport of Asian Meteorology, editedby C. P. Changand T. N. dust. Krishnamurti,235-270, Oxford UniversityPress, New York, 1987. Acknowledgments.We thankcooperating organizations Prospero,J. M., D. L. Savoie,R. T. Nees,R. A. Duceand J. in India(Aliga•h and Kashmir University), Pakistan (Water Merrill, Particulatesulfate and nitrate in the boundary and Power DevelopmentAgency), and Nepal (Tribhuvan layerover the north Pacific ocean, J. Geophys.Res. 90, University)for assistancein the field, andS. Whitlowfor 10,586-10,596. assistancein thelaboratory. The workin Indiaand China was Ramage,C. S., MonsoonMeteorology, AcademicPress, supportedby the NationalScience Foundation (USA), in New York, 1971. Pakistanby Wilfrid LauderUniversity and the International RoyalGeographic Society, The Mountains of CentralAsia, DevelopmentResearch Center (Canada), and in Nepalby the 1:3,000,000 map and Gazeteer,Macmillan, London, GeneralElectric Company Young Employees Nepal 1990 1987. Expedition(England). Savoie,D. L., J. M. Prosperoand E. S. Saltzman,Non-sea- salt sulfate and nitrate in trade wind aerosols at Barbados: References Evidencefor long-rangetransport, J. Geophys.Res 94, 5069-5080, 1989. Sequeira,R. andD. Kelkar, Geochemicalimplications of Alekseyev,B. A. et.al., GeographicBelts and Zonal Types summermonsoonal rainwater composition over India, J. ofLandscapes ofthe World. Scale 1: 15,000,000.School App. Meteor. 17, 1390-1396, 1978. ofGeography, Moscow State University, Moscow. 1988. Thompson,L. G., et al., Glacialstage ice-core records from Bartie,L. A. andJ. M. Hales,The spatial distribution of the subtropicalDunde Ice Cap, China,Annal. Glaciol. precipitationacidity and major ion wet deposition in 14, 288-297. NorthAmerica during 1980, Tellus 36B, 333•355, 1984. Wake, C. P., Glaciochemicalinvestigations as a tool for Barry,R. G. andR. J.Chorley, Atmosphere, Weather & determiningthe spatialand seasonal variation of snow Climate- 4th ed., Methuen,New York, 1982. accumulation in the Central Karakoram, Northern Belikova,T. V., V. N. Vasilenko,I. M. Nazarov,A. N. Pakistan,Annal. Glaciol. 13, 279-284, 1989. Pegoevand S. D. Fridman,Characteristics of background Wake, C. P., P. A. Mayewskiand M. J. Spencer,A review of sulfatepollution of thesnow cover on the territory of the centralAsian glaciochemicaldata, Annal. Glaciol. 14, USSR,Meteorologiya i Gidrologiya 9, 47-55,1984. 301-306, 1990. Chen,K. andJ. M. Bowler,Late Pleistocene evolution of salt Wake, C. P., P. A. Mayewski,P. Wang, Q. Yang,J. Han and lakesin theQaidam Basin, Qinghai Province, China, Z. Xie, Anthropogenicsulfate and Asian dust signals in Paleogeog.,Paleoclimat., Paleoecol. 54, 87-104, 1986. snow from Tien Shan, northwestChina, Annal. Glaciol. Dregne,H. E.,Surface materials ofdesert environments, in 16, 45-52, 1992. Desertsof theWorld, edited by W. G. McGinnies,B. J. Whitlow S., P. A. Mayewskiand J. Dibb, A comparisonof Goldmanand P. Paylore,286-377, University of Arizona majorchemical species seasonal concentration and Press,Tuscon, 1968. accumulation at the South Pole and Summit, Greenland, Galloway,J.N., D. Zhou,J. Xiong and G. E. Likens,Acid Atmos. Environ. 26A, 2045-2054, 1992. Rain:China, United States, and a remotearea, Science Yasunad, T., Seasonal weather variations in Khumbu Himal, 236, 1559-1562, 1987. Seppyo 38, 74-83, 1976. Harte,J., An investigationof acid precipitation in Qinghai Province,China, Atmos. Environ. 17, 403-408,1983. Paul A. Mayewski andCameron P. Wake, Glacier Hastenrath,S., Climateand Circulationin the Tropics,D. ResearchGroup, Institute for the Studyof Earth,Oceans and Reidel, Boston, 1985. Space,University of New Hampshire,Durham, New Inoue,J., Climateof KhumbuHimal, Seppyo 38, 66-73, Hampshire03824 1976. Li Zhongqin, Wang Ping and Xie Zichu, Lanzhou Junge,C. E. andR. T. Werby,The concentration ofchloride, Instituteof Glaciologyand Geocryology,Academia Sinica, sodium,potassium, calcium and sulfate in rainwater over Lanzhou 730000, China the UnitedStates, J. Meteor.15, 417-425,1958. Luo,H. andM. Yanai,The large-scale circulation and heat sourcesover the Tibetan Plateauand surroundingareas Received: March 23, 1993 duringthe early summer of 1979.Part I: Precipitationand Accepted: May 5, 1993