ticular, to John Splettstoesser of the Minnesota Geolog-ical Survey. We also thank Charles Swithinbank, DavidDrewry, and William Kosco and his colleagues in theUnited States Geological Survey, and the Naval aerialphotographic team.

Our field party consisted of Charles Swithinbank(through the courtesy of the British Antarctic Survey),Henry Brecher, James Fastook, William Pfeffer, MarkHyland, Geoffrey Lingham, and Terence Hughes. Ourwork was supported by NSF grant DPP 77-22204.

References

Denton, G. H., and H. W. Borns, Jr. 1974. Former groundedice sheets in the Ross Sea. Antarctic Journal of the United States,vol. 9,

p. 167.

Denton, G. H., H. W. Borns, Jr., M. G. Grosswald, M. Stuiver,

and R. L. Nichols, 1975. Glacial history of the Ross Sea.Antarctic Journal of the United States, vol. 10, pp. 160-64.

Mercer, J .

H. 1968. Glacial geology of the Reedy Glacier area,Antarctica. Bulletin of the Geological Society of America, vol. 79,pp. 471-86.

Mercer, J .

H. 1972, Some observations on the glacial geologyof the Beardmore Glacier area, Antarctica. In Antarctic Ge-ology and Geophysics, ed. R.

J . Adie, pp. 427-33. Oslo, Nor-way: Universitetsforlaget.

Stuiver, M., G. H. Denton, and T. Hughes. In press. Historyof the marine ice sheet in West Antarctica during the lastglaciation: a working hypothesis. In The Last Great Ice Sheets,ed. G. H. Denton and T. Hughes. New York: Wiley-Inter-science.

Swithinbank, C. W. 1963. Ice movement of valley glaciers flow-ing into the Ross Ice Shelf, Antarctica. Science, vol. 141, pp.523-24.

Microparticle deposition at SouthPole

E. MOSLEY-THOMPSON and L. G. THOMPSON

Institute of Polar StudiesThe Ohio State UniversityColumbus, Ohio 43210

The primary goal of this investigation was the detailedtnalysis of the particles within a 101-meter firn coreIrilled in 1974 at the Amundsen Scott (South Pole) Sta-;ion (90°S).

The antarctic ice sheet is an ideal site for investigatingemporal variations in the global background particulateiiass (Hogan, 1975) in that there are few local sources.In addition, because atmospheric motion over East Ant-srctica is predominantly a matter of subsidence (VanLoon and Williams, 1977) and surface outflow, the up-er tropospheric and stratospheric masses are probably

:he primary sources of near surface material. Therefore,my substantial increase in the concentration of particlesvithin the global atmosphere should be recorded in theIntarctic ice sheet.

A total of 6,218 samples representing 65 meters ofwater were analyzed for particle concentration and sizelistribution. Individual particles within selected sectionswere examined for morphology and elemental constit-ients using a scanning electron microscope and an x-raymergy dispersive system. The detailed sampling madet possible to detect annual variations in particle concen-ration (figure 1), and these were used to construct a)1 1-year time scale for the core. The time scale consti-.utes the longest record of its type from Antarctica. NetLccumulation does not exhibit a general trend over thentire 911-year interval, although the 90 years from A.D.

590 to 1680, a period characterized by the lowest global

PARTICLES WITH DIAMETERS k 0.63 jim (10)

0.020.040.060.080.0100.0120.0140.014.0

1889

Li-4TH1888

1887

1886 C12,400

________ 1885- 1884

15.04 sugges t ed as Krakatau (Aug. 26, 1883)

- -, 1883

= 1882-

15.51^- -::------4-9,250

18797

1878

Figure 1. Concentration of particles with diameters v 0.63micrometer per 500 ji 1 sample in a 2-meter section of theSouth Pole core for period including 18709 and 1880s. Key:C = average concentration over depth interval encom-passed by arrows;? = peak for which interpretation is un-

certain.

temperatures during the last millenium, exhibits con-sistently low rates.

The concentration of nonsoluble particles with di-ameters ' 0.63 micrometer per 500 j. 1 sample (figure2) increases substantially between A.D. 1450 and 1850,the years of the Little Ice Age. This period of reducedmean global temperatures also is marked by frequentand explosive volcanic activity (figure 2). The remark-able similarity between the profiles of nonsoluble par-ticle concentration and global volcanic activity suggeststhat the additional material is of volcanic origin. Theslopes of the particle size distribution (figure 2) espe-cially support this suggestion.

The average slope of the size distribution for the1950-56 strata is -3.2, very similar to the -3.2 (±0.3)measured for the total atmospheric aerosol at the SouthPole (Shaw, 1975). Gras and Laby (1979) report a re-duction in the size distribution slope of the total aerosol

91

SO SO SO SO5050iO50 soODInInIninininininin

=2

0

0

0

ICd

4 m

4I',F-

0in0

In

.5.0tinI')

0

00.0

C',

N

inN

0

0In

>0—o0.00

In

0

0

0

4

SO505050505050505050inininininIninInininin = 0

'CV JD9A

Figure 2. Concentration of particles with diameters 0.63 per 500 1 sample for period spanning A.D. 1056 to 1956. Columnshow, from left to right, the concentration of particles with diameters 0.63 ji per 500 .L 1 sample; the dust veil index (Dvi) fcthe southern hemisphere; the global Dvi; the annual average of the characteristic slope of the size distribution for each firsample; and the central England temperature profile (Lamb, 1965), often used to represent a global trend (GARP no. 16). KeA = period of intense volcanic activity, A. D. 1810-1840; 8 = period of volcanic activity, A. D. 1750-1770; and C = 1883 eruptlo

of Krakatau (Krakatoa). (The letters suggest correlative peaks between the concentration of particles and the DVI profiles.)

92

from -4 to -3 for their photoelectric data measured inthe stratosphere between 16 and 22 kilometers at lati-tude 34°S after the eruption of Fuego (14°13'N/91°W)in October 1974. Note that the slopes are lowest whenthe concentrations are highest in the South Pole core(figure 2).

On the basis of all the data (Mosley-Thompson andThompson, 1979), it appears that the South Pole snowstrata are recording the fluctuations in the concentrationof nonsoluble particles in the atmosphere, not only overthe southern hemisphere but over the entire globe.Thus, microparticle data from deep ice cores will pro-vide valuable information about the global particulatemass over many millennia.

This work has been supported by National ScienceFoundation grant DPP 76-07745.

ReferencesGras, J . L., and J . E. Laby, 1979. Southern hemisphere strat-

ospheric aerosol measurements. 2. Time variations in the

1974-1975 aerosol events.Journal of Geophysical Research, 84:303-07.

Hogan, A. W. 1975. Antarctic aerosols. Journal of Applied Me-teorology, 14: 550-59.

Lamb, H. H. 1965. The early medieval warm epoch and itssequel. Palaeogeography, Palaeoclimatology, Palaeoecology, 1:13-37.

Lamb, H. H. 1970. Volcanic dust in the atmosphere with achronology and assessment of its meteorological signifi-cance. Philosophical Transactions of the Royal Society of London,A266: 425-533.

Mosley-Thompson, E. M., and L. G. Thompson. In press. 911years of microparticle deposition at the South Pole. (Nature.)

Shaw, E. G. 1975. Climatic implications of central Antarcticaerosols. Antarctic Journal of the United States, 10: 188-89.

Van Loon, H., and J . Williams. 1977. The connection betweentrends of mean temperature and circulation at the surface:Part IV. Comparison of the surface changes in NorthernHemisphere with the upper air and with the Antarctic inwinter. Monthly Weather Review, 105: 636-47.

World Meteorological Organization. 1975. The Physical Basis ofClimate Modelling. GARP Publication Series, no. 16. Geneva.

Stable isotope and radio echosounding investigations of Taylor

Glacier, Victoria Land

D. J . DREWRY

Scott Polar Research InstituteCambridge UniversityCambridge, England

During January 1978 and January 1979, samples ofice and snow for stable isotope analyses were collectedfrom Taylor and neighboring glaciers in southern Vic-toria Land in order to evaluate and test ideas concerningice provenance and glacial geologic history of theMcMurdo Sound region.

As a result of detailed radio echo sounding (REs) inVictoria Land during 1974-75 (part of the InternationalAntarctic Glaciological Project [IAGPI study of past andpresent dynamics of the East Antarctic Ice Sheet), an icedome and ridge system was identified inland of TaylorGlacier (Drewry, in press). This ice divide separates themain drainage of the ice sheet into Mulock Glacier tothe south and Mawson Glacier to the north. Smaller,intervening glaciers (Skelton, Ferrar, Taylor, WrightUpper, Victoria Upper, Mackay, etc.) are supplied onlyby local ice originating within a few tens of kilometersof the Transantarctic Mountains.

This conclusion has important implications for glacialchronology of Victoria Land. Although advances andretreats of these minor glaciers have been considered toreflect fluctuations of the East Antarctic Ice Sheet (Den-ton, Armstrong, and Stuiver, 1971; Hendy et al. 1979),it would now appear that only very large-scale ice sheetevents (that is, several hundred meters of thickening)would be of sufficient magnitude to reverse or modify

local slopes and allow penetration of additional ice intothe mountains, and thus be mimicked by movement ofterminus of Taylor and other glaciers. The RES data,therefore, suggest that the minor glaciers are nourishedprincipally by local ice, have glaciological regimes con-trolled by nearby mountain climatic effects, and thus areinsensitive gages of small- to medium-scale fluctuationsof the East Antarctic Ice Sheet.

The objectives of our 1978-79 project were to deter-mine the contemporary isotopic composition of snow inthe catchment area of Taylor Glacier; the isotopic valuesfor old ice at the Taylor Glacier snout; and the regionalrelationship between mean b values and elevation andmean annual surface temperature. (Sample locations areshown in figure 1.)

Our samples in 1978 consisted of six 1-meter-long icecores (obtained using a Sipre corer) taken horizontallyfrom the lowermost 1.5 meters of the Taylor Glaciersnout at a 25-centimeter vertical spacing. Three otherice cores were taken in a similar manner from WrightLower Glacier. All of these samples were analysed by L.Merlivat of the Centre d'Etudes Nucléaires de Saclay(Gi f-su r- Yvette, France) by arrangement with C. Loriusof the Laboratoire de Glaciologie (Grenoble, France).

In 1979, we undertook a more ambitious program,including a radio echo sounding profile flown down thecenterline of Taylor Glacier from the ice dome to LakeBonney and further collection of samples for stable iso-tope analysis.

A 4-meter vertical snow core was obtained close to thesummit of the dome inland of Taylor Glacier. The meanannual accumulation in this vicinity is about 140 kilo-grams meter-' (Crary, 1963), so that the core shouldaverage about 12 years' precipitation = 437 kil-ograms meter-'). A pit 1.5 meters deep was excavatedon Wilson Piedmont Glacier near the source of ice sam-pled in 1978. Estimating eight annual layers gave anapproximate accumulation rate for this locality of 70kilograms meter -2 year'. However, it is possible that,judging from Chin's results (unpublished), which were

93