Located on the western coast of the Ross Sea, the McMurdo Dry Valleys are extremely cold, having air temperatures that range from -45 C during winter months to about 5 C during summer months. The valleys receive less than 10 centimeters (cm) of precipitation per year (Wharton 1991, pp. 1-51), most of which typically is lost to sublimation within a matter of hours. Yet within this harsh landscape, abundant microbial mats exist in glacial meltwater streams, which flow intermittently during the austral summers. For most of the year, the streams are dry, and the algal mats are essentially "freeze dried." From November to January, however, the streams flow for a period of 6-10 weeks. Algal mats are present in the main channels of these streams or in rivulets draining the hyporheic zone and become photosynthetically active within hours of becoming wet (Vincent and Howard-Williams 1986). The algal mats consist of cyanobacteria--primarily Nostoc and filamentous Oscillatoria, whereas nematodes, tardigrades, and rotifers dominate invertebrate populations associated with the algae.

Taylor Valley, located in the McMurdo Dry Valleys, is the focus of much of the field research of the McMurdo Long-Term Ecological Research (LTER) program. Several Taylor Valley streams were chosen to investigate the spatial distribution of algal species composition and biomass. In all, 141 samples were collected from 12 different streams during January 1994 (ongoing LTER study involving the streams in Taylor Valley). To illustrate the approach and findings of this study, results from only one stream will be discussed in this report. Specifically, 10 samples from Green Creek, a short [approximately 1,500-meter (m)] stream flowing from Canada Glacier into Lake Fryxell, will be used to illustrate the use of a small-scale contour map in plotting algal species distribution relative to their position in the stream channel.

In the section of Green Creek chosen for analysis, two varieties of microbial mats were visibly present--"orange algae" and "black algae". Five samples of each variety were collected from the streambed using a number 13 core borer (having a diameter of 17 millimeters and an area of 227 square centimeters) and were preserved in formalin. Using a Nikon Diaphot phase contrast microscope, samples were examined in the laboratory to determine species compostion and abundance. Each subsample was placed in a 2 milliliter settling chamber (Utermohl 1958) and seven random fields (250 square micrometers) were examined at x400 magnification. Within each field, algae were identified to genus and species where possible (excluding diatoms, which were counted but not identified). Percentage of cover was measured for all specimens in each field, the seven fields were tallied, and percentage of the total algal biomass was determined for each taxa. Invertebrate counts were estimated by an examination of the entire slide and were noted on a subjective scale of rare, moderate, and abundant (table).

In general, "orange algae" were found in the main flow zone of the stream, whereas "black algae" were found on the outer edge of the flow zone or in the wetted zone (figure 1). (See Broady 1982, for a related study involving Canada Stream, Taylor Valley.) Specifically, "orange" samples 1-5 all were collected within 2 meters of the stream's thalweg (the down-channel course of the greatest cross-sectional depth within a stream channel). On the other hand, "black" samples 1, 2, 4, and 5 all were collected from minimal flow regions at the stream channel's edge, and "black" sample 3 was collected from a backwater eddy, also less susceptible to faster stream flow. The "orange algae" were usually dominated by filamentous species of Oscillatoriaceae, and the "black" algae primarily consisted of Nostoc (figure 2). "Orange" samples were composed of a wide diversity of algal species, whereas "black" samples were composed of over 90 percent Nostoc--a significant contrast between the spatially different samples. In addition, the percentage of Oscillatoria subproboscidea in "orange" mats appears to be inversely proportional to the percentage of Oscillatoria irrigua, but factors controlling this relation remain unknown.

Several hypotheses have been devised to explain observed patterns of species distribution across the stream channel. Foremost, Oscillatoria and Phormidium spp. generally are long and filamentous, making them more hydrodynamically suitable for life in the faster flowing main channel. Likewise, Oscillatoria and Phormidium spp. often intertwine into thin felts and tightly encrust themselves to the streambed, allowing these epilithic filaments to persist in rapidly flowing water. In contrast, colonies of Nostoc mature into large, mucilaginous colonies whose shape and consistency are more adapted to life on moist ground bordering the main channel. Stream flows in Taylor Valley can vary as much as tenfold in one day depending on air temperatures and insulating conditions (McKnight and Tate, Antarctic Journal, in this issue). Therefore, Nostoc colonies are normally submersed for only a few hours a day or not at all. Although more susceptible to drying effects from sun and wind, Nostoc in the wetted zone are not exposed to the abrasive effects of sediment transport occurring in the main channel.

In brief, Green Creek is a highly dynamic environment with numerous algal species spatially segregated across the channel. The stream environment varies greatly during the flow period, and many questions about habitat requirements of the algal mats remain unanswered. Nonetheless, this report illustrates the usefulness of small-scale stream contour maps in determining algal species distribution relative to their position in the stream channel. Continued research involving stream contour maps, in conjunction with water-quality and nutrient data, will provide a better understanding of the distribution and abundance of Taylor Valley algal mats and their overall importance to the stream ecosystems.

We thank Thomas L. Johnson for his assistance in the production of the stream contour map. This reserach was produced in cooperation with the National Science Foundation, U.S. Antarctic Program (grant OPP 92-11773). The use of trade names in this report is for identification only and does not constitute endorsement by the U.S. Geological Survey.

REFERENCES.

Broady, P.A. 1982. Taxonomy and ecology of algae in a freshwater stream in Taylor Valley, Victoria Land, Archiv fur Hydrobiologie (Supplementband 63, Algological Studies, 32), 331-349.

McKnight, D.M., and C.M. Tate. 1995. Algal mat distribution in glacial meltwater streams in Taylor Valley, southern Victoria Land, Antarctica. Antarctic Journal of the U.S., 30(5).

Utermohl, H. 1958. Toward the improvement of the quantitative phytoplankton method. Mitteilungen-Internationale Vereiningung fur Limnologie, 9, 1-38. (In German).

Vincent, W.F., and C. Howard-Williams. 1986. Antarctic stream ecosystems: Physiological ecology of a blue-green algal epilithon. Freshwater Biology, 16, 219-233.

Wharton, R.A. (Ed.). 1991. McMurdo Dry Valleys: A cold desert ecosystem (Report of a National Science Foundation Workshop held at the Institute of Ecosystem Studies, Millbrook, New York October 5-7). Reno, Nevada: Desert Research Institute.