Lake levels have been increasing throughout the McMurdo Dry Valleys since detailed records began in 1960 (Chinn 1993). Further, there is also evidence that the level of Lake Bonney has been rising since the dry valley lakes were first discovered by Scott's party in 1903 (Chinn 1993). Lakes in the dry valleys are fed by glacial meltwater streams that flow for 6 to 9 weeks during the austral summer, typically during December and January. The rising lake levels, therefore, are caused by increased streamflow associated with longer periods of temperatures above freezing and other climatic factors (Chinn 1993). The ultimate cause of the warming trend in climate in the dry valleys is not known.
In addition to being the main source of water to the lakes, glacial meltwater streams are also important sources of inorganic solutes to the lakes (Green et al. 1989). Solutes in the streams come from leaching of marine aerosols and calcite, as well as from primary weathering reactions occurring within the streambed alluvium (Green et al. 1989; McKnight and Andrews in press). Calcite is abundant in the dry valleys because of calcite deposition in Lake Washburn, which filled most of the Taylor Valley until about 10,000 years before present. The abundance of marine aerosol is dependent upon the proximity to the coast. Therefore, the relative ratio of sodium (Na) to calcium (Ca) as major cations is expected to decrease with increasing distance from the coast.
Lake Hoare is a closed-basin lake located in Taylor Valley with its eastern boundary determined by the Canada Glacier (figure); it has been rising since records began in 1972 with a net rise of 1.35 meters (m) (Wharton et al. 1992; Chinn 1993). The major streams flowing into Lake Hoare receive drainage from the Canada Glacier. Lake Chad is located west of Lake Hoare and has a surface area about a tenth of that of Lake Hoare. Lake Chad receives drainage from the Suess Glacier and can, in turn, drain into Lake Haore. Since 1956, flow from Lake Chad to Lake Hoare has been seen on aerial photographs of the area, specifically on 5 December 1956, 1 January 1958, 12 January 1975, 18 January 1981, and 20 December 1982 (Wharton personal communication).
In late January 1992, we observed from a helicopter fly-over that Lake Chad appeared to have merged with Lake Hoare. On 30 January and 1 February 1992, we studied the narrows area between the two lakes by making water-level measuremetns and collecting samples for analysis of major ions. A sample was also obtained above the lake bottom from a slush zone at a depth of 6.5 m at a site near the center of Lake Chad. The water samples were filtered through 0.4-micrometer (um) Nucleopore filters using an Antlia filtration unit; samples for cation analysis were acidified with Ultrex nitric acid. Cations were analyzed by inductively coupled plasma spectroscopy and anion analysis by ion chromatography.
On 30 January 1992, we surveyed the water level in the moat areas from the north shore of both Lake Hoare and Lake Chad relative to the New Zealand benchmark NZARPBMT15 (76.95 m above mean sea level) located between the two lakes. The water level for both lakes was 74.21 m above mean sea level. We interpret these measurements as showing that Lake Hoare has risen sufficiently such that it is now at the same level as Lake Chad. Depending on the inflow from meltwater streams to either lake in the future, the hydrologic connection between the two lakes could go in either direction. Whereas, in the past when the surface of Lake Chad was higher than Lake Hoare, the hydrologic connection, when it occurred, was always from Lake Chad to Lake Hoare.
The results of water sample analyses are presented in the table. The results show significant variation in chemistry. First, the sample of the bottom water of Lake Chad had relatively high solute concentrations, with Na and sulfate (SO4) much greater than Ca and chloride (Cl). The results for the Lake Chad outlet and the two sites in the narrows are all similar and had Ca:Na atomic ratios between 1.3 and 1.0. At the Lake Hoare inlet, the Na concentrations were comparable to those in the narrows, but the Ca concentration was about half resulting in a lesser Ca:Na ratio of 0.6, The other major divalent cations, magnesium (Mg) and strontium (Sr), and the anions SO4 and Cl showed the same pattern as Ca, being greater in the narrows than in the Lake Hoare inlet. The observation that the water in the narrows, the transition zone between the lakes, is most like Lake Chad water suggests that the average direction of flow is from Lake Chad to Lake Hoare. During January 1992, there were other indications of substantial rise in the level of Lake Hoare such as the encroachment of the lake waters near a field camp established in 1987.
The water chemistry in the open water "moat" areas of the lakes at the end of the streamflow period is probably controlled by the chemistry of the streamflow and mixing with the lakewater. The ice cover is about 3.5 to 4 m in Lake Hoare whereas in Lake Chad, the ice appeared to extend to 6.5 m, close to the lake bottom. Therefore, mixing with lake water may be more significant for meltwater coming into either of the two basins in Lake Hoare. The greater Na:Ca ratio in the Lake Hoare inlet area may possibly reflect differences in the relative importance of sources of solutes, such as leaching of marine aerosols, dissolution of calcite, and primary weathering reactions.
The merging of the two lakes will cause the meltwater from the Suess Glacier to interact more directly and rapidly with the water in the upper layer just below the ice in Lake Hoare. Both hydrologic and geochemical changes can be envisioned. First, the more direct hydrologic transport from the "Lake Chad" lobe could influence the circulation in the main lake, possibly causing a west-to-east flowpattern in the zone just below the ice cover. Such a flow from the west lobe to the east lobe has been observed in Lake Bonney (Priscu personnal communication). Greater velocities under the ice could have several consequences. For example, the major cation composition could shift in the zone just below the ice. Also transport of algal and bacterial species from Lake Chad to Lake Hoare could occur. Another biological consequence may be enhancement of growth for diatoms, which require some mixing to avoid sinking. In 1987, we found very low abundance of several diatoms (for example, Cyclotella sp. and Navicula muticopsis) just below the ice cover (depths of 4.1 and 5.0 m), and no diatoms were found below 7.5 m. Greater abundances and diversity of diatoms may occur if there were a layer below the ice with flow going from the "Lake Chad" lobe to Lake Hoare.
It is possible that the changes associated with the merging of the two lakes would leave a record of an "abrupt" change of some kind in the lake sediments, such as the appearance of particular diatom species in the sedimentary record. These observations point out that distinct or abrupt events affecting the physical system can occur during the 10- to 100-year periods of gradual, but steady, climatic change. The possibility of such physical events might be considered in the paleolimnological study of lake sediments to elucidate conditions in the past.
We thank R. Harnish and V.C. Stephens for their assistance in the analysis of major ions. This research was supported by the National Science Foundation grant OPP 88-17113.
References
Aiken, G.R., D.M. McKnight, R.L. Wershaw, and L. Miller. 1991. Evidence for the diffusion of aquatic fulvic acids from the sediments of Lake Fryxell, Antarctica. In R.A. Baker (Ed.), Organic substances and sediments in water. Ann Arbor: Lewis Publishers.
Chinn, T.J. 1993. Physical hydrology of the dry valley lakes. In W.J. Green, E.I. Friedman (Eds.), Physical and biogeochemical processes in antarctic lakes. (Antarctic Research Series, Vol. 39.) Washington, D.C.: American Geophysical Union.
Green, W.J., T.J. Gardner, T.G. Ferdelman, M.P. Angle, L.C. Varner, and P. Nixon. 1989. Geochemical processes in the Lake Fryxell basin (Victoria Land, Antarctica). Hydrobiologia, 172, 129-148.
McKnight, D.M., Aiken, G.R., Andrews, E.D., Bowles, E.C., and Harnish, R.A. 1993. Dissolved organic material in Dry Valley lakes: A comparison of Lake Fryxell, Lake Hoare, and Lake Vanda. In W.J. Green, E.I. Friedman (Eds.), Physical and biogeochemical processes in antarctic lakes. (Antarctic Research Series, Vol. 39.) Washington, D.C.: American Geophysical Union.
McKnight, D.M., and E.D. Andrews. In press. Hydrologic and geochemical processes at the stream-lake interface in a permanently ice-covered lake in the McMurdo Dry Valleys, Antarctica. Verhandlungen International Verein Limnologie.
Wharton, R.A., Jr. 1993. Personal communication.
Wharton, R.A., Jr., C.P. McKay, G.D. Clow, D.T. Andersen, G.M. Simmons Jr., and F.G. Love. 1992. Changes in ice cover thickness and lake level of Lake Hoare, Antarctica: Implication for local climatic change. Journal of Geophysical Research, 97(C3), 3503-3513.