ture: 50 percent plagioclase, 25 percent quartz, 20 percentbioiite, and 5 percent apatite, chlorite, and zircon.

510-1—Biotite granite: 10 percent plagioclase partly re-placed by microcline and conspicuously sericitized, 70 percentK-feldspar that is slightly cataclastic, and 20 percent biotitethat is partly chloritized and replaced by epidote micro-veining, accessory zircon.

510-3—Pegmatitic granite: 65 percent K-feldspar that ispartially albitized and sometimes contains relics of more basicplagioclase, 3 percent quartz, and minor plagioclase that issericitized.

5I0-4--Biotite-garnet gneiss: 25 percent plagioclase thatis highly sericitized and albitized at junctions with K-feldspars, 35 percent microcline, 23 percent quartz, 7 percentbiotite, and 10 percent garnet.

552, 552-2, and 552-3: Pickering Nunatak. A migmatizedsequence of granulite facies rocks with a slight degree ofdiaphthoresis. Hypersthene plagiogneiss (552) contains bandsof leucogranite (552-2) and both are cut by veins ofyounger granites (552-3).

552—Biotite-hypersthene plagiogneiss: 52 percent plagio-clase (An45), 15 percent quartz in granulated bands, 18percent hypersthene, 13 percent biotite after hypersthene,2 percent opaques, and accessory apatite and zircon.

522-2--Cataclastic leucogranite: 40 percent microcline,25 percent plagioclase, 35 percent quartz, and rare muscoviteafter microcline.

552-3--Slightly cataclastic leucogranite: 40 percentmicrocline, 40 percent quartz, 20 percent plagioclase, andaccessory biotite and zircon.

Subterranean flow into Lake Bonney

BARRON L. WEAND, RICHARD D. FORTNER, andROBERT C. HOEHN

Civil Engineering Department

BRUCE C. PARKERBiology Department

Virginia Polytechnic Institute and State UniversityBlacksburg, Virginia 24061

Lake Bonney (77°43'S. 162'23'E.), a meromictic,permanently ice-covered lake in Taylor Valley, south-ern Victoria Land, has been the subject of numerousliinnological investigations (Armitage and House,1962; Angino and Armitage, 1963; Angino et al.,1964; Goldman, 1964; Yamagata et al., 1967). Since1972 researchers from Virginia Polytechnic Instituteaid State University have been making an intensiveeffort to model the lake's unique ecosystem. Emphasisto date has been on in situ monitoring of seasonalvariations in the lake's major chemical and biologicalparameters (Parker et al., 1973; Parker et al., inpress; Hoehn et al., in press; Craig et al., 1974).

Lake Bonney is a nearly closed ecosystem. It ispermanently stratified with a chemocline beginningat about 10 meters and a hypersaline monimolimnionextending to the maximal depth (34 meters). Thelake is covered by approximately 3.5 meters of per-manent ice. During the austral summer, the extremeeastern end of the lake becomes ice-free and a narrow(less than 10 meters) moat forms along the lake'sentire periphery. Glacial meltwater only enters the

lake during a 4- to 6-week period (in December andJanuary) of each year.

The present report is primarily based on chemicalchanges in Lake Bonney's water and on measurementsand estimates of glacial meltwater flows into thelake during the 1973-1974 summer. The latter, how-ever, were hindered by problems of siltation andunexpectedly high water flows (Hoehn et al., inpress; Hoehn et al., 1974).

Based on an observed increase in lake level (1.03meters) during the 1973-1974 summer, we calculatethat the lake volume increased by 3.26 million cubicmeters, or by about 5 percent. Surface flow to thelake was from the following glaciers: Sollas, LaCroix,Matterhorn, Hughes, Calkin, Rhone, and Taylor(figure 1). The last of these is by far the most im-portant; we visually estimate it to represent 74 per-cent of the total surface flow to the lake. Althoughwe were not able to precisely measure the flow fromeach of these sources, our best estimate for the totalsurface flow is 1.32 million cubic meters. Angino(1964) estimated surface flow into Lake Bonney to

January/February 1975 15

Seasonal variations in sodium, magnesium, potassium, and calcium at site 1, Lake Bonney, during the 1973-1974 austral summer.*

Depth Sodium Magnesium Potassium Calcium(meters)IIIIIIIIIIIIIIIIIIIIIIII

1944

102220640

1,380940

3,600

6895

123312423920700

1,500

4.027596

6.0760570

8.01,7802,050

10.03,6004,700

12.56,2808,600

15.011,60016,200

20.026,00011,300

26.043,00048,000

34037

32076

1,790190

-630

10,9002,400

26,1009,300

41,0008,60035,000 18,500

1840

107135

532460

1,310-

2 1 8503,8006,600 12,3007,100 15,500

22,000 13,300

1032

3764

133100

265-

412490

1,3901,600930 4,000

3,5502,600

4586

95103

162169

360-

380300

9001,060

7001,430

1,5801,160

*Sampling dates: I, November 25, 1973; II, December 12, 1973; III, January 19, 1974. All concentrations are inper liter.

be 295,000 cubic meters during the 1961-1962 sum-mer. It should be noted that the 1973-1974 summerwas unusually warm and the flow was apparentlymuch greater than in recent years. We estimate thatsurface flow during the 1973-1974 summer accountsfor only 41 percent of the increase in lake volume,leaving 1.94 million cubic meters to be from othersources. Because of these and other data collectedduring the 1972-1973 and 1973-1974 summers, wenow believe that there is subterranean flow into LakeBonney.

Angino and Armitage (1963) suggested possilinflow of thermal waters to explain Lake Bonnehigh chloride levels. A later study (Angino et1964) found excessive loss of water (by ablationice) over gain by inflow and suggested inflowdepth might compensate for the discrepancy.

On November 23, 1973, a vigorous effervescerwas noted at our sampling site 1 (figure 1). A simi]phenomenon was reported previously by Koob aLeister (1965) in conjunction with drilling sampliholes near our site 1. They supposed the gas to

S

t

Figure 1. Significant sourcesof runoff to Lake Bonney inrelation to sampling sites.

16 ANTARCTIC JOURNAL

-202468

E

CL 20a,

30

FIgure 2. Temperature anddissolved oxygen profiles in-dicative of bottom fluctua-tions at site 1A during the

1972-1973 austral summer. Temperature, °C

oxygen produced photosynthetically in the mixolim-nion. In our experience, however, the effervescenceoccurred in a drill hole that had been open forabout 3 weeks. The odorless gas evolved is presumablycarbon dioxide or oxygen, although we did not deter-mine its chemical nature. It would be possible fordissolved gases carried in any subterranean water tobe released upon entering the hypersaline monimolim-nion of Lake Bonney due to decreased solubility ofthe gas in the saline water (American Public HealthAssociation, 1971). It is noteworthy that during thesummer we found a dramatic increase in oxygenconcentration below the chemocline. Dissolved oxygenat the bottom of site 1 (26 meters), for example,was 0.8 milligrams per liter on November 12, 1973,but 13.7 milligrams per liter on December 24, 1973.A similar change (shown in figure 2) was observedat the bottom of the lake during the 1972-1973summer at site 1A (Parker et al., 1973). The dramaticincrease that occurred in dissolved oxygen on January1, 1973, was correlated with an increase from —2° to0°C. at the bottom of site 1A (31.5 meters). Duringthe 1973-1974 summer we found, in addition to anincrease in dissolved oxygen at the bottom of site 1,a substantial increase in sulfate from 2,250 milligramsper liter on November 12, 1973, to 4,930 milligramsper liter on January 15, 1974 (figure 3); a decreasein pH from 7.5 to 5.7 also was recorded during thisperiod.

The fluctuations of sodium, magnesium, potassium,

and calcium show similar patterns at site 1 (table).There is a decided seasonal decrease at the bottomtogether with increases at depths 10 meters abovethe bottom.

In addition to the chemical changes noted above,from December P1973 to January 1974 there weresignificant biological changes in both the mixolimnionand monimolimnion. In several instances, increasesin the concentrations of various parameters in themixolimnion were accompanied by decreased concen-trations at several depths in the monimolimnion andthe chemocline. On December 26, 1973, for example,bacteria and yeasts were not detected below 15 meters;by 1 to 2 orders of magnitude, however, they weremore numerous in the mixolimnion than they wereon the previous sampling date 2 weeks earlier. Surfacemeltwater flow could not account for the increases inheterotrophic organisms at this time, although thetransport of bacteria and yeast to the lake by the meltstreams at a later date was readily apparent by in-creases in the total plate count from less than 20colonies per milliliter to more than several hundredper milliliter (Hoehn et al., in press; Parker et al.,in press). Primary productivity also reached a maxi-mum on December 26, 1973. This increased produc-tivity occurred while inorganic nutrient levels alsowere increasing in the mixolimnion. There weresudden increases in inorganic nitrogen (primarilyammonia) and less dramatic increases in orthophos-phorus at this time. The contribution of inorganic

January/February 1975 17

0I2345SULFATE, g/l

Figure 3. Seasonal variationsin sulfate concentration atsite 1 during the 1973.1974

austral summer.

ANTARCTIC JOURNAL

nutrients by surface melt streams occurred too latein the season to account for these increases in nitrogenand phosphorus.

The aforementioned changes are compatible withthe hypothesis that there is significant, although inter-mittent, subterranean inflow of fresh meltwater intoLake Bonney at depths below 20 meters. Such inflowof fresh, oxygenated water would result in the fol-lowing: dilution effects of the principal cations atdepth; an increase in dissolved oxygen levels untiloxygen was consumed by inorganic and/or organicreactions; a rise in sulfate concentration and decreasein pH possibly due to the oxidation of sulfides,

H2S + 202 —4 SO42 + 211

biological and chemical changes in the mixolimniondue to upwelling of the inflow waters through thedeeper lake waters that are relatively enriched incertain nutrients and organisms.

Understanding such inflow and its effects on the

chemical and biological parameters of this lake willbe important to our ecosystem model.

We would like to thank Gary Crouch for hismeasurements of surface runoff during the 1973-1974summer, Mary Halliburton for her assistance in metalanalysis by atomic absorption and emission spec-troscopy, and James Craig (Geology Department,Virginia Polytechnic Institute and State University),for his review of this manuscript. This research wassupported by National Science Foundation grant GV-35171.

References

American Public Health Association. 1971. Standard Meth-ods for the Examination of Water and Wastewater. NewYork, American Public Health Association. 13th edition:480.

0

5

10E

15

2020

25

18

Angno, Ernest E., and Kenneth B. Armitage. 1963. A geo-cFemical study of lakes Bonney and Vanda, Victoria Land,Aitarctica. Journal of Geology, 71: 89-95.

Angno, Ernest E., Kenneth B. Armitage, and Jerry C. Tash.164. Physicochemical limnology of Lake Bonney, Antarc-tia. Limnology and Oceanography, 9: 207-217.

Arrntage, Kenneth B., and Hugh B. House. 1962. A limno-loica1 reconnaissance in the area of McMurdo Sound,Attarctica. Limnology and Oceanography, 7: 36-41.

Crag, James R., Richard D. Fortner, and Barron L. Weand.174. Halite and hydrohalite from Lake Bonney, TaylorValley, Antarctica. Geology, 2(8): 389-390.

Goliman, C. R. 1964. Primary productivity studies in antarc-ti lakes. In: Biologie Antarctique (Carrick, R., M. W.Foidgate, and J . Prevost, editors). Paris, Hermann.2)1-299.

Hochn, Robert C., et al. In press. Nitrogen and phosphorusaailability to plankton and benthic communities in Lakelbnney, southern Victoria Land, Antarctica. ThirdSCAR/IUBS Symposium on Antarctic Biology, Proceed-izgs. Washington, D.C., National Academy of Sciences.

Hoehn, Robert C., Bruce C. Parker, and Robert A. Paterson.1974. Toward an ecological model of Lake Bonney. Ant-arctic Journal of the U.S., IX(6) : 297-300.

Koob, Derry D., and Geoffrey L. Leister. 1972. Primaryproductivity and associated physical, chemical, and bio-logical characteristics of Lake Bonney: a perennially ice-covered lake in Antarctica. Antarctic Research Series, 20:51-68.

Parker, Bruce C., Robert C. Hoehn, and Robert A. Paterson.1973. Ecological model for Lake Bonney, southern Vic-toria Land, Antarctica. Antarctic Journal of the U.S.,VIII(4): 214-216.

Parker, Bruce C., et al. In press. Changes in dissolvedorganic matter, photosynthetic production and microbialcommunity composition in Lake Bonney, south VictoriaLand, Antarctica. Third SCAR/IUBS Symposium onAntarctic Biology, Proceedings. Washington, D.C., Na-tional Academy of Sciences.

Yamagata, N., Tetsuya Toni, and Sadas Murata. 1967.Report of Japanese summer parties in the dry valleys,1963-65: part V, chemical composition of lake waters.Antarctic Record, 29: 2339-2361.

Peru's Quelccaya Ice Cap:glaciological and glacial geological studies, 1974

J. H. MERCER and L. G. THOMPSONInstitute of Polar Studies

The Ohio State UniversityColumbus, Ohio 43210

C. MARANGUNICDepartment of Geology

University of ChileSantiago, Chile

JOHN RICKERUniversity of British ColumbiaVancouver, British Columbia

Canada

Long ice cores have been obtained in recent yearsfrom Greenland and Antarctica, and shorter coreshave been retrieved from other high-latitude glaciers.It became evident that a stratigraphic record froma glacier in the tropics would be needed to test cer-tain assumptions made in the interpretation of polarice cores and to aid in an interhemispheric correlationof polar ice cores by providing data from an inter-mediate location.

Most glaciers in the tropics are on rugged moun-tains or steep volcanic peaks. Only two tropical icecaps are known to exist, both in the Southern Hemis-phere: in Irian Jaya (formerly Dutch New Guinea),at 4 0 S., the North Wall Firn (elevation: 4,800meters) covers about 4 square kilometers, and inPeru, at 14 0 S., the Quelccaya Ice Cap (elevation:5,500 meters) covers about 70 square kilometers. TheQuelccaya Ice Cap is larger and thicker than the

January/February 1975 19