Paula Adkins 421 Michigan Sandpoint, ID 83864 Phone:(702) 673-7492 (DRI) (800) 688-8606 (ASA) Home:(208)263-2974 Email:padkins@maxey.dri.edu		Prof. Yo Chin 327 Menden Hall Laboratory 125 South Oval Mall Ohio State University Columbus, OH 43210 Phone:(614) 292-7688 Fax:(614) 292-6953 Email:yo@geology.ohio-state.edu
Dr. Alex Blum U.S. Geological Survey 3215 Marine St. Boulder, CO 80303 Phone:(303) 541-3027 Fax:(303) 447-2505 Email:aeblum@usgs.gov		Prof. Robert Collier Western Nevada Community College 3288 College Drive Carson City, NV 89703 Phone:(702) 887-3101 Home:(702) 883-0270 Fax:(702) 887-3175 Email:collier@scs.unr.edu
Alexandra C. Brown INSTAAR and Department of Civil, Environmental and Arch. Engineering 1560 30th Street Campus Box 450 Boulder, CO 80309-0450 Phone:(303) 492-4703 or 5811 Fax:(303) 492-6388 Home:(303) 444-0590 Email:alexcb@snobear.colorado.edu		Gayle L. Dana Biological Sciences Center Desert Research Institute 7010 Dandini Blvd. (89512) Reno, NV 89506-0220 Phone:(702) 674-7538 or 7539 Home:(702) 674-0108 Fax:(702) 673-7485 Email:gdana@maxey.dri.edu
Melody B. Burkins 6105 Fairchild Hall Department of Earth Sciences Dartmouth College Hanover, NH 03755 Phone:(603) 646-1612 Home:(802) 649-5733 Fax:(603) 646-1682 Email:Melody.Brown.Burkins@dartmouth.edu		Dr. Robert E. Davis Cold Regions Research and Engineering Laboratory Hanover, NH Phone:(603) 646-4219 Email:bert@hanover-crrel.army.mil

	Page
Session I - Paleolimnology and Isotope Geochemistry..............................	16
Session II - Stream Chemistry and Hydrology......................................	19
Session III - Physical and Chemical Limnology....................................	22
Session IV - Algal Mats..........................................................	26
Session V - Plankton.............................................................	29
Session VI - Surface Energy......................................................	32
Session VII - Soil Processes and Ecology.........................................	37
Lunch Discussion - Soil Ecosystem and Stream Ecosystem Linkages: Planned Experiments for the 1997-98 Field Season.................................	40

Doran, Peter T. and Robert A. Wharton Jr., Biological Sciences Center, Desert Research Institute, Reno, NV (e-mail: pdoran@maxey.dri.edu and wharton@maxey.dri.edu)

A strong reservoir effect in the terrestrial Antarctic carbon pathway has long been cited as making carbon dating problematic, yet very little research has been carried out on the nature and extent of the effect. Recent investigations have shown that stream microbial mat and some surface lake water DIC date modern, while other surface lake waters carry a relict signal. The controlling factor seems to be the mode of stream input. Direct input from glaciers introduces old carbon, but if water travels through streams, it is allowed to equilibrate with modern carbon dioxide before entering the lake. Once the radiocarbon is in the lake, strong stratification can create DIC bottom water ages in excess of 10,000 yr B.P. Due to these phenomena, radiocarbon dating is a viable technique for lake edge deposits, and lake bottom deposits where a correction to the sediment surface age is possible. Paleolake deposits can not be reliably dated using radiocarbon dating alone because the age of the reservoir correction (i.e. accounting for the initial carbon reservoir, plus the age of the bottom water) can not be determined.

Diatoms assemblages in surficial sediments, sediment cores, sediment traps, and inflowing streams of perennially ice-covered Lake Hoare, South Victoria Land, Antarctica were examined to determine the distribution of diatom taxa, and to ascertain if diatom species composition has changed over time. Lake Hoare is a closed-basin lake with an area of 1.8 km2, maximum depth of 34 m, and mean depth of 14 m, although lake level has been rising at a rate of 0.09 m yr-1 in recent decades. The lake has an unusual regime of sediment deposition: coarse grained sediments accumulate on the ice surface and are deposited episodically on the lake bottom. Benthic microbial mats are covered in situ by the coarse episodic deposits, and the new surfaces are recolonized. Ice cover prevents wind-induced mixing, creating the unique depositional environment in which sediment cores record the history of a particular site, rather than a lake-wide integration. Shallow-water (<1 m) diatom assemblages (Stauroneis anceps, Navicula mol esta, Diadesmis contenta var. parallela, Navicula peraustralis) were distinct from mid-depth (4-16 m) assemblages (Diadesmis contenta, Luticola muticopsis fo. reducta, Stauroneis anceps, Diadesmis contenta var. parallela, Luticola murrayi) and deep-water (26-31 m) assemblages (Luticola murrayi, Luticola muticopsis fo. reducta, Navicula molesta). Analysis of a sediment core (30 cm long, from 11 m water depth) from Lake Hoare revealed two abrupt changes in diatom assemblages. The upper section of the sediment core contained the greatest biomass of benthic microbial mat, as well as the greatest total abundance and diversity of diatoms. Relative abundances of diatoms in this section are similar to the surficial samples from mid-depths. An intermediate zone contained less organic material and lower densities of diatoms. The bottom section of core contained the least amount of microbial mat and organic material, and the lowest density of diatoms. The dominant process influencing species composition and abunda nce of diatom assemblages in the benthic microbial mats is episodic deposition of coarse sediment from the ice surface.

Lake Fryxell and Hoare in Taylor Valley, Southern Victoria Land, together with other Antarctic lakes, are unique in having a perennial ice cover. This ice cover limits gas exchange between the atmosphere and the lake water, and causes a very stable stratification of the lakes. We analyzed a series of water samples from profiles of these lakes and their tributaries for d13C of the DIC in order to qualify the carbon flux from the streams into the lakes, and to investigate the carbon cycling within the lakes. Isotopic values in the uppermost waters (d13C = +1.3 to 5.3 in Lake Hoare, +0.4 to +3.03 in Lake Fryxell) are close to the carbon isotope values encountered in the streams feeding Lake Fryxell, but distinctively heavier in Lake Hoare (stream d13C = -2.3 to +1.4). These ratios are much heavier than ratios found in the moat that forms around the lakes in January/February (d13C = -10.1). In the oxic photic zones of the lakes photosynthesis clearly influences the isotopic composition with layers of hi gh productivity having enriched carbon isotope signatures (d13C = +2.7 to +6.1). In both lakes, the isotopic values become lighter with depth and reach minima of -3.2 and -4.0 at the bottoms of Lake Fryxell and Hoare, respectively. These are caused by the microbial remineralization of isotopically light organic carbon. We present DIC flux calculations that help interpret the isotopic distribution. For example, in Lake Fryxell slightly higher CO2 recharge via streams, and a substantially larger diffusion of CO2 from deeper waters causes the lighter observed isotopic ratios. Differences in flow regimes and stream morphologies of the tributaries also greatly influence the carbon budgets of the basins.

The McMurdo Dry Valleys in southern Victoria Land, Antarctica (76 30' - 78 30'S, 160-164E) rank along with the most extreme deserts in the world, and represent the coldest and driest of the Long Term Ecological Research sites. Numerous glaciers in the region terminate in a closed basin and have seasonal runoff that contribute to three perennially ice covered lakes in the area. Patterns of isotopes of surface ice on all glaciers in Taylor Valley do not behave in a similar manner. Deuterium isotope data of glacier ice from the lower Canada glacier indicate a distinct flow pattern of ice with a symmetric ~ 50 per mile difference between the glacier center and the margins. This may be a result of flowing ice originating at different elevations, reflecting a modern isotopic elevation gradient for precipitation. Alternatively, this pattern could also represent ice from the late Holocene cooling trend which has been found in the Taylor Dome isotope record and in lacustrine data from McMurdo Dry Valleys. Isotopes along a centerline transect on Taylor Glacier indicate a pattern of isotopically lighter, probably Glacial age ice, above more enriched ice at the terminus which may shed light on Taylor Dome isotopic interpretation.

Conovitz, Peter A., Diane M. McKnight, Lee H. MacDonald, Andrew G. Fountain, Harry R. House, Department of Earth Resources, Colorado State University, Fort Collins, CO (e-mail: pcono@meeker.CNR.ColoState.edu and leemac@cnr.colostate.edu), INSTAAR, University of Colorado, Boulder, CO (e-mail: mcknight@snobear.colorado.edu), U.S. Geological Survey, Denver, CO (fountaina@pdx.edu), U.S. Geological Survey, Madison, WI (e-mail: hrhouse@dwimdn.er.usgs.gov), Currently on leave to the Department of Geology, Portland State University, Portland, OR

In the McMurdo Dry Valleys, glacial meltwater streams are a critical linkage between the glaciers and the lakes in the valley bottoms. This paper analyzes the physiographic characteristics and six years of discharge data from five streams in order to better characterize the dynamic inputs into Lake Fryxell, a closed basin in Taylor Valley. These feeder streams typically flow only for six to eight weeks during the summer, and streamflow is highly variable on an interannual as well as daily basis. During low flow years, the shorter streams contributed a higher proportion of the total annual inflow into the lake; this pattern may reflect the greater losses to wetting ht hyporheic zone. Comparisons of the period of direct sun on the glacier faces with the time of peak flow suggested that solar position and melt from the glacier faces are the dominant controls on the diurnal fluctuations in streamflow. An analysis of streamflow recession showed considerable variability between streams and in some cases, over time. For example, recession coefficients for Canada Stream, a short stream with an incised channel, were fairly invariant with streamflow. In contrast, the recession coefficients for Lost Seal Stream, an unconfined, low gradient stream, increased significantly with increasing discharge. These observations lead to hypotheses for the control of streamflow dynamics in the McMurdo Dry Valleys by climate, solar position, and geomorphic factors.

Despite the cold, dry climate in Taylor Valley (TV), Antarctica, there is evidence for chemical weathering in the TV streams. Thus the chemistry of the TV streams is influenced not only by glacial composition, salt dissolution, and atmospheric precipitation, but also by rock weathering. This study attempts to determine the relative influences of each rock type on the stream chemistry by using the ratios of both the major and minor alkali metals and alkaline earth elements. Cations within each family have the same charge which allows for substitution of the heavier and less abundant ones for the lighter ones that are commonly found in primary minerals. These measured rations have been compared among TV streams and streams draining monolithologic watersheds in the United States and elsewhere, as well as, to seawater and crustal values. From the K/Rb and Sr/Ba data, the TV streams lie between the seawater and monolithologic watersheds data, indicating both marine and rock weathering influences. The Rb/Cs a nd Sr/Ba data indicate that TV streams include solutes of marine origin and weathering from one or more of the following: sandstone, carbonate and metamorphic rocks. Also, the aqueous ratios are compared to ratios in stream bed sediments and unweathered rocks, in order to determine the ions available for weathering and the ions retained in secondary minerals. Since the heavier cations are less mobile, the heavy/light ratio decreases from rock to sediment (weathered bedrock) to water. According to XRD analysis, quartz, albite, actinolite, kaolinite, and illite are present in the TV stream sediments. In addition, sediment surfaces were viewed on a SEM to observe evidence of chemical weathering. Besides the lithology of the watershed, such factors as the presence/absence of active biological processes and physical characteristics of the stream have been taken into account to determine the effects on stream chemistry.

Silicate weathering rates were calculated along Von Guerard stream during the Antarctic summer using three synoptic studies at differing flow conditions. The stream is analogous to a saturated flow column reactor with pH varying from 7 to 8. Dilute surface flow from a dry bottom glacier flows ~5.2 km to Lake Fryxell, interacting along the channel with saturated gravels in the hyporheic zone, with no higher plants and with very limited biological activity. The hyporheic zone volume is well defined by the thawed region in the permafrost. Side slopes thaw to <30 cm and sublimation precipitation, so there is no runoff, and negligible contribution of solutes from the side slopes to the stream. The stream gravels are composed of sub-equal quantities of i) granites, ii) intermediate intrusive and metamorphic rocks, and iii) diabase and basalt. BET surface areas (SA) were measured on 10 size fractions from <63 mm to 1.4 mm, and yielded surface roughnesses of 230 to 580, which did not vary systematically with grain size. This indicates that internal porosity is the major contribution to SA. XRD analysis of the <2mm size fraction indicates minor amounts of kaolinite and primary silicates, but predominately amorphous material. This suggests that in this cold, ephemeral stream the formation of secondary clay minerals is inhibited.

Elemental fluxes, calculated from changes in solution chemistry along the stream reach, are used to quantify weathering. Si is the best single indicator of weathering reactions. At high flow conditions, when the stream interacts strongly with the hyporheic zone, Si concentrations increase linearly downstream from 128 to 167 mM (R2=0.94), yielding a silicate weathering rate of 8.26x10-3 mmol Si m-1 of stream reach. Using the total hyporheic zone volume and stream flow yields a specific weathering rate (mol Si/cm2/sec) of 10 15.3 using BET SA and 10 12.8 using geometric SA. At lower flow, the Si weathering rates are up to 60% slower, and less consistent along the stream reach. This decrease reflects reduced circulation through the gravels and recharge to the stream from the hyporheic zone.

Mass balance calculations between potential reactant and product phases were used to attempt to trace the sources and sinks of solutes. However, the large number and poorly constrained chemistry of the solid phases make the results poorly constrained. Nevertheless, the evolving solution chemistry is consistent with probable weathering reactions, and increases in K and Mg down the stream reach can only be accounted for by silicate weathering reactions occurring at appreciable rates.

The measured weathering rates based on Si concentrations are very rapid, greater than any reported watershed study or experimental dissolution rates for plagioclase, and similar to the most rapid experimental dissolution rates for hornblende and augite. These rapid rates suggest: i) saturated aqueous environments, with greater mineral surface/water contact, may be significantly more reactive than unsaturated soil environments; and ii) biological activity does not dramatically accelerate silicate weathering rates in watersheds, since this essentially abotic weathering is more rapid than has been reported for warmer, forested watersheds.

Lyons, W. Berry, Kathy A. Welch, Klaus Neumann, Jeffrey K. Toxey, Robyn McArthur, Changela Williams, Diane M. McKnight, and Daryl Moorhead, Department of Geology, University of Alabama, Tuscaloosa, AL (e-mail: Lyons - bylons@wgs.geo.ua.edu, Welch - kwelch@ualvm.ua.edu, and Neumann - kneumann@wgs.geo.ua.edu), INSTAAR, University of Colorado, Boulder CO (e-mail: mcknight@snobear.colorado.edu), Department of Biology, Texas Tech University, Lubbock, TX (e-mail: tudlm@ttacs.ttu.edu)

Aquatic systems in the Taylor Valley, Antarctica have characteristics of other desert aquatic systems in that the amount of water is limited and highly variable during the year. As part of the McMurdo Dry Valleys LTER, we have examined the major element chemistry of the three largest lake basins in order to investigate the geochemical continuum and the geochemical processes occurring within the Taylor Valley. During the summer, meltwater is generated from the glaciers and flows through streams to perennially ice-covered lakes in the valley bottom. As water moves through the system, solute concentrations increase by orders of magnitude. The glacier data suggest that some amount of salt is recycled from the soils and blown by winds onto the glaciers. Spatial differences in glacier chemistry have been observed and these, along with characteristics of the glacial meltwater streams, result in differences in stream chemistry within the valley. Dissolution of evaporite salts within the stream channels, as w ell as the weathering of Si minerals appear to be significant geochemical processes especially in the longer streams. The differences in modern day stream chemistry would lead to different chemical evolutionary pathways for the different lakes. High interannual variability of stream flow has also been observed which leads to differences in the amount of fresh water and solutes entering into the lakes each season. In addition, seasonal chemical changes occur within the lakes due to the inflow of freshwater and biological activity. For example, changes in calcite saturation in the lakes have been observed through the austral summer period. Based on our work, it appears that long-term systematic monitoring of stream and lake hydrology and chemistry is needed in order to quantitatively evaluate water and solute balances for the lakes, as well as to understand lake dynamics.

The McMurdo Dry Valleys harbor the most extreme and unique lake systems on Earth. Studies of these lakes have been undertaken in the past to define these extreme ecosystems, and to monitor the impact of global change on this sensitive environment. Until recently, there was believed to be three main lake types in the dry valleys (in order of prominence): 1) perennially ice-covered lakes (i.e. with 3-6 m of lake ice over a water body of variable depth and salinity), 2) ice block lakes (frozen to their bed), and 3) ice free lakes (small brine pools). However, during a pilot study performed in November 1995 in collaboration with the McMurdo Long Term Ecological Research project, two supposedly ice block lakes were surveyed using ground-penetrating radar (GPR), and the results indicated the presence of saline water bodies beneath up to 19 m of lake ice. In October of 1996 drilling confirmed the presence of a perennial NaCl brine (in an ice matrix) at 15.8 m depth in Lake Vida, existing at -11.6 C. The pilot study failed to penetrate deep enough into the Lake Vida ice cover to sample the saline water body beneath. The preliminary results indicate that the brine within the ice matrix must be approximately 600 ppt salinity (about 17 times more concentrated than seawater). Microbial biomass indicates that eubacterial and microalgal cells (primarily filamentous cyanobacteria) are associated with sedimentary material throughout the ice matrix. Assays performed on ice core meltwater demonstrate that the populations of both heterotrophic and autotrophic microbes were metabolically active (measured via the incorporation of radio-labeled CO2, thymidine and leucine) once thawed, suggesting that the ice-bound microbial populations are capable of growth when liquid water becomes available within the permanent ice environment.

We present high-resolution measurements of conductivity and temperature made from January 1990 to December 1993 in the east and west lobes of Lake Bonney and in Lakes Vanda, Hoare, Fryxell, Joyce, and Miers. These measurements were used to calculate profiles of density and stability, and thereby infer mechanisms and strengths of mixing in the water columns of the lakes. Transects along the length of Lake Bonney allowed estimates of horizontal exchanges in and between the two lobes of that lake and help to explain some of the characteristics of single profiles measured in other lakes. Stratification in all the lakes is controlled mainly by concentration of dissolved solids (salinity), with temperature exerting such a minor influence as to act virtually as a passive tracer. An exception is in the upper two-thirds of Lake Vanda and at the bottom of Lake Miers, where solar heating in the presence of weak salinity gradients gives rise to thermohaline convection. The distinctive and relatively invariant shapes of the density profiles in the different lakes is due to distinctive distributions of salts in the water columns of these lakes, distributions that can only be explained in terms of geochemical processes acting over time scales much longer than the annual overturning cycle that dominates patterns of stratification and mixing in temperate, freshwater lakes. Temperatures in the McMurdo Dry Valleys lakes, in contrast to salinities, do respond to changes in weather, climate, and water levels on a seasonal and annual basis, although to a much smaller extent than in temperate lakes. Stability reaches extremely high levels in the chemoclines of the two lobes of Lake Bonney, being slightly lower in the bottom waters of Lake Vanda. Stabilities in Lakes Fryxell and Joyce, although still very high in comparison with freshwater lakes, are much lower than in Bonney and Vanda. Maximum stabilities in Lakes Hoare and Miers are similar to those found in the summer thermoclines of freshwater lakes. With the exception of thermoha line convection cells in Lake Vanda and Lake Miers, our measurements do not support the presence of turbulent diffusion in the main bodies of the lakes; however, profiles did document mechanically generated turbulence just below the ice in Lake Miers (probably associated with the meltwater stream through-flow in that lake, the only lake with a stream outlet) and much weaker turbulence in the narrows connecting the two lobes of Lake Bonney (probably associated with the exchange flows between these basins).

In this work, we utilize chlorofluorocarbon (CFCs) concentrations profiles from Lakes Hoare and Fryxell in the McMurdo Dry Valley to determine the extent of deep vertical mixing occurring over the last 50 years. Near the ice-water interface, CFC concentrations in both lakes are well above saturation, in accordance with atmospheric gas supersaturations. Evidence of mixing throughout the water column at Lake Hoare was confirmed by the presence of CFCs throughout the water column and suggests mixing times of 20-30 years. In Lake Fryxell, CFC-11, CFC-12 and CFC-113 were found in the upper water column, however degradation of CFC-11 and CFC-12 in the anoxic bottom waters appears to be occurring, with only CFC-113 present in these bottom waters. The presence of CFC-113 in the bottom waters, in conjunction with previous work detecting tritium in these waters, strongly argues for convective mixing to also be occurring in Lake Fryxell. The evidence for deep mixing in these lakes may be an important, phenomenon in the limnology of these perennially ice-covered lakes.

Major ion chemistry can be used to examine the physical and chemical connections between the Taylor Valley streams and lakes, as well as the examine physical processes within the lakes. The perennial ice covers of the dry valley lakes inhibit mixing due to winds or temperature driven turnover. However, density driven mixing can occur given sufficient concentration of the surface waters under the ice. As new lake ice is added to the bottom of the ice cover during the winter, slats are effectively excluded from the ice matrix and are concentrated in the lake surface water. In general, the inflowing stream water has much lower concentrations of dissolved solids, by more than an order of magnitude, than the lake surface water. However, since the austral summer of 93-94, the amount of new water being added to the lakes has been less than the predicted water loss by sublimation. The result has been a general increase in surface water salinity. In Lake Hoare, which has the least chemical stratification of the lakes, our data from the 96-97 field season show evidence of density driven mixing and the loss of stratification down to 8 meters. Later during that filed season, stratification was reestablished in Lake Hoare as fresh water was added to the surface. Our data suggest that the source of this fresh water may be a combination of stream inflow and lake ice meltwater because concentrations of this new surface water were even lower than the stream water entering the lake. Lake Bonney and Lake Fryxell did not experience the same type of density driven mixing. This is likely due to the fact that they are more highly stratified and more saline than Lake Hoare. However, all the lakes exhibit the same trend of increasing salinity over the period from 1993-1997 in the top few meters of the water column. During each summer season the surface water salinity decreases due to the addition of new fresh water, but the year to year trends show an overall increase in surface water salinity due to relatively low stream fl ows over this period. If the trend continues, we might expect an increase in density driven mixing in the other lakes, as well. Continued monitoring of major ion chemistry may provide a useful tool for interpretation of physical, chemical, and biological processes within the lakes.

C.M. Tate, D.M. McKnight, A. Alger, G. Shupe, S. Spaulding, Water Resources Division, U.S. Geological Survey, Denver, CO (e-mail: Tate - Ctate@LTERnet.edu), Institute for Arctic and Alpine Research, University of Colorado, Boulder, CO (e-mail: mcknight@snobear.colorado.edu), Geologic Division, U.S. Geological Survey, Reston, VA, Department of Invertebrate Zoology and Geology; California Academy of Sciences, San Francisco, CA (e-mail: spauldin@CAS.calacademy.org)

The abundance and distribution of algal mats were studied in three streams flowing into Lake Fryxell, located in lower Taylor Valley. Algal mats were most abundant at sites which have moderate gradients and streambeds composed of large cobbles arranged in a flat stone pavement through periglacial processes. Algal abundance was less at high gradient and deltaic sites. Most of the length of the three streams can be characterized as large cobble and the total chlorophyll a in each stream was estimated from measurements made at representative sites. Four algal mat types were used to characterized the stream biota. Black-, orange-, and green-colored algal mat types occurred at most sites, but red-colored meats occurred in only one of the streams. At all sites, black-colored mats were found near the channel margins and green-colored mats were found on the underside of rocks in the main channel. Orange- and red-colored mats occurred in flowing water habitats, either in the main channel or in rivulets draini ng the hyporheic zone at the stream margins. Thus, similarities in physical characteristics of the stream habitat appeared to determine the occurrence of the different algal mats rather than differences in water quality. The species composition of the different mat types was consistent among sites. The black-colored algal mats were dominated by Nostoc sp, with a low average evenness of 0,13 +/- 0.07. The green-colored algal mates were also essentially unialgal composed chiefly of Prasiola calophylla or P. crispa, and having a low average evenness of 0.17 +/- 0.10. The orange- and red-colored mats also had a high degree of intrasite heterogeneity.

Perennial mats of filamentous cyanobacteria occur in glacial meltwater streams in the McMurdo Dry Valleys, Antarctica. To learn about long-term microbial survival in Antarctica, we routed meltwater to a relict channel for which the last recorded period of sustained flow was 1969. The relict algal mats in the channel began growing soon after becoming wetted. Transfer experiments showed that the relict mats grew more rapidly than mats in streams with regular summer flow, probably because of greater solute and nutrient concentrations. In the dry valleys, algal mats survive in potential stream habitats through cryogenic preservation, such that stream and lake ecosystems respond rapidly to geomorphologic shifts.

A sediment addition experiment has been designed to assess rates of benthic microbial mat growth as well as mechanisms by which sediment recolonization occurs in ice-covered lakes. Sediment enclosures and simulated sediment dumps were deployed in Lake Hoare in January 1997, with the intent of resurveying and resampling at 2 and 4-yr intervals. Sediment used in the experiment is identical to sediments introduced to the lake benthos during actual dump events. The experiment involves a total of nine treatments plus controls. Six of these treatments have been set up in triplicate and three in quintuplicate, and involve the use of clear acrylic sediment pots, cylinders and discs. Additions include 1) natural sediment, 2) ashed (i.e., OM-free) sediment and 3) unashed but sterile (i.e., autoclaved) sediment. Treatments will allow us to assess the importance of the sediment "seed bank" of naturally-occurring microbes, sediment dump thickness, and colonizer source (overlying water column, buried benthic material , horizontal migration) to the recolonization process. Triplicate cores were also collected at the initiation of the experiment in order to measure pre-treatment biomass, physicochemical characteristics and community composition of in situ sediment and mat material. Data on initial OM content (by AFDW), chlorophyll-a, carbon and nitrogen content of sectioned cores and added sediments will be presented. It is expected that this experiment will yield interesting and useful results on both the rate and principal modes of biomass recolonization of lake-deposited sediments.

New and published information on production of microbial communities in streams of the McMurdo dry valleys is reviewed. The dominant community in many of these streams is a thick cohesive cyanobacterial mat. Light/photosynthesis relationships of microbial mat communities from a range of streams tend to show a surprising degree of convergence. Gross rate of photosynthesis typically approach an upper limit of 4 mg C cm-2 h-1 at ambient temperature, and community light saturation intensities are almost always below incident irradiance during the period when streams are flowing. Net and gross photosynthesis increase with increasing temperature, and our analysis supports previous views that temperature is the prime determinant of rate of net production in these communities. There is a tendency for higher respiration rates in thicker mat communities, resulting in these mats tending toward a zero net gas exchange, i.e. where gross photosynthesis @ respiration. Accumulation of new material on exposed surfaces i s slow, and most communities are clearly at least 3-4 years old. We argue that the development of high biomass communities, which are balanced or near-balanced with respect to gas exchange, is possible due to the lack of disturbance within areas of these streams, a high rate of overwinter survival, and the constancy of growth conditions during the flow period.

Edwards, Robert Lyman and John C. Priscu, Biology Department, Montana State University, Bozeman, MT (e-mail: Edwards - ubire@gemini.oscs.montana.edu and Priscu - Jpriscu@LTERnet.edu)

The McMurdo Dry Valley (MDV) lakes are physically stable environments in which depth stratified phytoplankton assemblages develop over the austral summer. Phytoplankton at specific depths grow under relatively constant temperature, light and nutrient conditions. These conditions, while relatively constant at specific depths within a lake, vary widely between depths and among lakes. We examined the partitioning of 14C-bicarbonate into the protein, polysaccharide, lipid and low-molecular weight macromolecules of phytoplankton in the MDV lakes. The patterns of carbon partitioning within the various phytoplankton assemblages were measured during in situ experiments run concurrently with baseline 14C primary productivity measurements. We will discuss the effects of season, depth and lake on these patterns and relate them to in situ temperature, light and nutrient conditions. We will also discuss the results of laboratory incubation experiments in which photosynthate partitioning was examined under varying conditions of light and nutrients, and over a time course of 48 hrs.

Using automated overwinter sampling devices, we collected preserved phytoplankton samples from multiple depths in Lake Fryxell, a permanently ice-covered lake in southern Victoria Land, Antarctica. Photosynthetic algae (i.e. algae possessing chloroplasts) are maintained in a stable water column throughout winter darkness. The algal taxa overwinter in different ways, in a species specific manner. Typical vegetative cells were the most abundant form for all species found in the water column. Populations of one chlorophyte, Stichococcus sp., and two cryptophyte species increased during winter. We interpret the increase in algal population size as evidence of wintertime heterotrophic growth, and mixotrophic behavior in the contest of the entire year. For two chlorophyte species some portion of the population has distinct morphology, e.g. akinetes for Chlamydomonas subcaudata and cells containing a large amount of starch or other storage material for Chlorella sp.. During winter, vegetative cells of the m ost abundant species of cyanobacteria, Phormidium angustissimum, occurred a the depth of the summertime maximum and at depths below the oxycline, which may represent a false bottom. Other than this false bottom and the absence of diatoms, settling did not appear to influence the overwintering algal community.

Lake Hoare and Lake Fryxell were frequently sampled for protozooplankton between Oct. 1996 and Jan. 1997. In common with other Antarctic lakes both Lake Hoare and Lake Fryxell were dominated by microbial plankton. In Lake Fryxell the most abundant phototrophic nanofagellate (PNAN) was a cryptophyte species, which formed a stable deep maximum (max. 6000 ml-1) above the chemocline throughout the summer season. In both lakes heterotrophic nanoflagellates (HNAN) were sparse (mean; 110 ml-1 Fryxell, 82 ml-1 Hoare), relative to PNAN (mean: 1250 ml-1 Fryxell, 578 ml-1 Hoare). The success of cryptophytes in these poorly illuminated environments is partially due to their ability to ingest bacteria. Ingestion rates were measured using fluorescently labeled bacteria (FLB) and ranged between 2.2 to 0.6 bacteria cryptophyte-1 h-1.

Ciliates were extremely abundant with a maximum density of 60 ml-1 occurring at the chemocline in Lake Fryxell. Ciliate diversity was high (over 20 species) compared to other Antarctic lakes, with many of the species exhibiting vertical zonation (Holophrya sp., Askenasia sp., small Monodinium sp.). A number of the ciliate species appear to be mixotrophic either through sequestration of plastids)from phytoplankton prey (Strobidium viride) or through the harboring of endosymbiotic algae (Bursaria sp.). Mixotrophy appears to be an important survival strategy in these lakes.

During my January, 1997 stay at Lake Hoare as a Teachers Experiencing Antarctica (TEA) grantee, I had the opportunity to examine rotifer (Philodina spp.) and tardigrade (Hypsibius sp.) populations living on top of algal mats in nearshore areas and ponds adjacent Lake Hoare. These organisms comprise the vast majority of multicellular biomass in the lake. As organisms at the top of their food chain in the lake ecosystem, it seems relevant to know something about their diets and feeding rates. The question being asked is: What are typical feeding rates of rotifers and tardigrades? The approach taken to answer this question involved the use of bacterial and phytoflagellate-sized fluorescent latex microspheres (FLM). Results obtained using the FLM method to estimate grazing rates will be presented. Problems with the method will be discussed and ideas for changes in methodology are welcome.

Perennially ice-covered Antarctic lakes exemplify microbially-dominated ecosystems existing at the extremes of conditions found on our planet. In four of the lakes found in Taylor Valley (Hoare, Fryxell, Bonney and Joyce), planktonic, extracellular virus-like particles (VLPs) are plentiful. In two of these lakes, large, possibly novel icosahedral forms have been observed by transmission electron microscopy (TEM), and in one lake thus far examined, viruses appear to have potentially high production potential. We summarize data on water column VLP distributions in four lakes collected during the 1995/96 and 1996/97 field seasons. Based on our data, we suggest that virally-mediated mortality may be a major biotic factor regulating the abundance of many of the microbes in these extreme environments. In addition, antarctic lakes may constitute a large reservoir for undiscovered viruses possessing novel characteristics.

Takacs, Cristina D. and John C. Priscu, Montana State University, Bozeman, MT (e-mail: Takacs - ubijpct@gemini.oscs.montana.edu and Priscu - Jpriscu@LTERnet.edu)

Research of the microbial ecology of the McMurdo Dry Valley lakes has primarily concentrated on phototrophs; relatively little is known about the bacterioplankton. Production by aquatic heterotrophic bacteria is an important link among detritus, dissolved organic matter, and higher trophic levels in many aquatic systems. Bacterial numbers in the lakes of the dry valleys range from 105-107 cells ml-1 which is comparable to temperate fresh and marine systems. Bacteria represent a substantive proportion of water column biomass in these lakes comprising 30-60% of microplankton biomass. Bacterial production and cell numbers were measured 3-5 times within four Antarctic seasons in Lakes Bonney, Hoare, and Fryxell, including the winterspring transition during one year. Seasonal trends and yearly variation in activity, biomass, specific activity, and production integrated throughout the water column will be discussed in relation to primary productivity, nutrients, and the potential role of grazing. Our study i s the first to describe the seasonal and annual variations in bacterial activity and numbers, revealing the dynamics of this component of the McMurdo Dry Valley lake ecosystem.

Dana, Gayle L., Scott S. Tyler, Robert E. Davis, and Karen Lewis, Biological Sciences Center, Desert Research Institute, Reno, NV (e-mail: gdana@maxey.dri.edu), Water Resources Center, Desert Research Institute, Reno, NV (e-mail: scott@maxey.dri.edu), Cold Regions Research and Engineering Laboratory, Hanover, NH (e-mail: bert@crrel41.crrel.usace.army.mil), Institute of Arctic and Alpine Research, University of Colorado, Boulder, CO (e-mail: kjl@tintin.colorado.edu)

Surface energy exchange is important to ecosystems in McMurdo Dry Valleys, Antarctica because it drives glacier and lake ice ablation, glacial stream discharge, soil heat and moisture transfer and other processes. Different energy transfer processes have different effects on ecosystem properties, such as species distribution and productivity. Surface energy exchange consists of radiation, sensible and latent heat exchange and ground heat transfer. Latent and ground heat fluxes are coupled with moisture transfer. We measured the surface energy balance over glaciers, lakes and soils in the dry valleys for short time periods (a few days) during midsummer of 1995-1996 using eddy correlation techniques. Net radiation was the dominant energy source for all seven sites. Energy losses were partitioned differently among sites. Sensible heat exchange was the dominant heat-loss factor from soil surfaces, while latent fluxes dominated heat loss over glacier and lake ice. Daily sublimation and evaporation rates were calculated from latent heat flux. Sublimation from glacier and lake ice ranged from 1.3 - 2.5 mm per day. The daily evaporation rate from soils was much smaller, about 0.35 mm per day. Magnitudes and partitioning of energy fluxes will likely change for dry valley glaciers, lakes and soils under different climate regimes. Point measurements provide a fundamental understanding of energy transfer processes occurring over the different s urfaces. However, effects of climate change scenarios may best be explored by spatially distributing energy and mass transfer models. Gridded incoming solar radiation (Dana et al. in press) is an example of one of the first steps toward this goal. An initial approach would be to segment land cover and terrain to apply a one-dimensional energy balance model (Davis et al., 1995, e.g.). Validation of model results should tie in both point measurements from eddy correlation techniques and meteorological stations, and spatial data from remote sensing.

Solar radiation is an important driving force for hydrological and biological systems in the dry valleys, influencing sublimation and melting of the glaciers, heating of the soils and air, and providing energy for photosynthesis by the microbial communities in the streams, soils, and perennially ice-covered lakes. We analyzed two years of solar radiation data from eleven meteorological stations positioned on glaciers, lake shores, and lake ice in Taylor, Wright, and Victoria Valleys. Average annual incoming solar radiation ranged from 84 to 117 W m-2 during 1994 and 1995. We attribute differences among stations primarily to terrain effects, but coastal cloudiness and orographic effects may also be factors.

Because of the importance of terrain to solar radiation patterns, we applied a topographic solar radiation model to Taylor Valley, using in situ pyranometer data to drive the model. Considerable topographic variability in solar radiation occurs over the region, even averaged over a monthly time scale, with north facing slopes receiving more energy than south facing slopes. In the valley bottom, differences in incident radiation were discerned among lakes, with Lake Fryxell receiving uniform amounts of energy while Lakes Hoare and Bonney received less energy along their northern shores due to terrain shading. Hourly radiation maps and pyranometer data illustrate that the terminus of the glaciers receive higher levels of solar radiation than their surface, but this intense illumination is of short duration, occurring only when the sun directly strikes the cliff face.

Future solar radiation modeling efforts will include the incorporation of multiple pyranometer data, solar radiation inputs from satellite data, as well as satellite-based estimates of net radiation.

Glaciers are the principal source of water to the perennially ice-covered lakes in the McMurdo Dry Valleys, Antarctica. During the last 20 years, dry valley lake levels have risen and lake-ice thickness has decreased in some cases. This implies a change in energy fluxes of the glaciers and water flow from them. Energy-induced changes in meltwater generation from the glaciers will change dry valley ecosystems through their effects on water and sediment flow, and solute and dissolved gas transport. Energy and hydrological relationships will ultimately be used to drive ecological models being prepared as part of the McMurdo Dry Valleys Long Term Ecological Research (LTER) project. The goal is to determine lake and stream ecosystem structure and function in the context of climate change.

In this study, NOAA AVHRR data were used to characterize the spatial and temporal distributions of conditions over the major land surface features in the dry valleys (snow, ice, soils and rock). Basic derived AVHRR data include surface temperatures and reflectance and the results of unsupervised spectral mixture analysis. Several energy based indices for estimating the generation of glacial meltwater were developed using the derived satellite data in conjunction with ground-based streamflow and glacial mass balance measurements. We compare satellite-derived melt indices with those calculated from degree-day models as a first step toward improving the multispectral products. This comparison shows the promise of using satellite-derived indices to predict glacial melt. More rigorous comparison and development will be possible with distributed energy balance models.

During the austral summers of 1995-96 and 1996-97 a portable, scanning-grating spectroradiometer was utilized to measure solar spectral irradiance at a range of depths in Lake Hoare. This device rapidly scans the range of wavelengths between 280 and 340 nm (which includes the complete spectrum of UV-B radiation) and is effectively used both in air and under water. In 1995-96, data were obtained down to a depth of 1.5 m in the ice. In the following year, scans were made at several locations under the full 4.5 to 5 m thick ice column. The use of the underwater spectroradiometer by SCUBA divers permitted avoidance of hole effects and rapid assessment of spatial variability in UV penetration. Results of this investigation and its relevance to potentially destructive threshold photon absorption energies will be discussed.

The surface energy balance was calculated to estimate sublimation and melt on the surface and terminus of Canada Glacier in Taylor Valley, Antarctica during the 1994-95 and 1995-96 austral summers. Our results indicate that sublimation accounted for roughly 80% of the observed 1994-95 summer ablation and 42% of the observed 1995-96 summer ablation on the surface of the glacier. Sublimation of the terminus cliffs appears to be less significant than sublimation on the glacier surface, probably accounting for at most 10-15% of the measured ablation. Based on these results, both surface and terminus cliff melt were calculated and compared with gauged flow in the glacial streams. We found that while the terminus cliffs represent only 2% of the total ablation zone, they account for 10 to 40% of the total meltwater runoff. Given our current instrumentation, we can estimate meltwater discharge from the glacier with an accuracy of 20%.

Strong spatial variations in glacier mass balance exist in Taylor Valley. The clearest example of this change is the rapidly rising equilibrium line, which divides the zones of net snow accumulation from net ice ablation. The meteorological factors that control the mass balance in Taylor Valley include air temperature, wind speed, solar radiation, and precipitation. The most important seem to be the latter two. Solar radiation increases up valley (away from the coast) and precipitation decreases. Both are related to cloudiness. Although one may expect this trend with distance from the ocean, which is the source of water vapor, the trend is intensified by the presence of Nussbaum Riegel, a 800 m bedrock obstruction in the center of Taylor Valley. The Riegel blocks moisture-bearing clouds from penetrating the valley into the Bonney Basin. Thus, snow occurs more frequently on the glaciers in the Fryxell Basin and less frequently in the Bonney Basin. Snowfall increases the albedo, decreases absorbed solar radiation and in turn decreases meltwater production. Therefore, the comparisons between streams in Fryxell Basin should be more variable than in the Bonney Basin.

Dana, Gayle L., Scott S. Tyler, and Robert A. Wharton, Jr., Biological Sciences Center, Desert Research Institute, Reno, NV (e-mail: Dana - gdana@maxey.dri.edu and Wharton - wharton@maxey.dri.edu), University of Nevada, Reno, P.O. Box 60220, Reno NV 89506.

Perennial ice covers influence lake ecosystems via their effects on physical properties such as spectral distribution and penetration of light, gas exchange between the water column and the atmosphere, wind-generated currents, and sediment deposition. Recent thinning of ice covers of numerous lakes in the McMurdo Dry Valleys, Antarctica have prompted interest in better understanding the processes contributing to this decline.

During the 1995-1996 austral summer we measured the surface energy balance of three lakes in Taylor Valley, Antarctica, using eddy correlation techniques. Here we will present the magnitudes and relative contributions of radiative, sensible and latent heat fluxes at the lake ice surface, which are the primary contributors to ablation of these ice covers. Differences in the energy balance of the ice cover surface among the three lakes will be discussed in terms of climatic regime, regional location, and ice surface conditions.

Virginia, Ross A., Ho Mengchi, Diana Wall Freckman, and Laura E. Powers, Dartmouth College, Hanover, NH (e-mail: Ross.A.Virginia@dartmouth.edu), Colorado State University, Fort Collins, CO (e-mail: freckman@nrel.colorado.edu), Eastern Mennonite University, Harrisonburg, VA

Desert soils are characterized by high spatial variability in chemical and biological properties . In warm deserts, spatial variation in soils is related to plant distribution and the activity of soil biota that are localized beneath plants. In contrast, Antarctic desert soils lack the dominating effects of higher plants and biotic activities are low. How does the soil spatial variability of the Dry Valleys compare with other deserts? We report variation in soil properties across a range of spatial scales in Taylor Valley as part of an effort to understand how the distribution, diversity, and function of soil invertebrate communities and soils are related. At large spatial scales, we can discern differences between the three LTER sites (Lakes Fryxell, Hoare, Bonney). Variation also exists within and between sorted polygons, the most obvious visible feature of the Dry Valley landscape. Across Taylor Valley, total soil C and N concentrations are highest near Lake Hoare. At the scale of the polygon, C a nd N concentrations tend to be higher near the cracks that define the borders of the polygons. The range of C and N variation found across large distances in Taylor Valley can also be found within much smaller sampling units. These results point to the importance of considering and understanding spatial variation in soils in the design of long-term experiments and for extrapolation of results to large-scale units (landscape to valley).

The source of soil organic matter in the Dry Valleys of the McMurdo Sound Region has long seemed problematic given the absence of higher plants in this arid ecosystem. The presence of visible accumulations of microbial mat material in dry valley lakes has lead to an LTER hypothesis that aeolian transport of this organic material may be an important source to dry valley soils. We examined the sources and distribution of organic matter from six elevational transects extending along a 30km section of Taylor Valley (approximately 163E, 77.35S), from the head of the valley to the Ross Sea. The d13C and d15N of soil were determined to identify potential sources of organic matter (i.e. marine, lacustrine, and/or terrestrial) to these soils and to map source distribution in the valley. Results suggest that the primary source for soil organic matter is not the wind-transport of organic material accumulating in present-day dry valley lakes. Instead, our data indicate that the primary source of SOM in Taylor Dry valley may be heavily influenced by (1) paleo-organic matter deposits entrained in the glacial tills and ancient lacustrine sediments lining the valley floor and (2) in-situ accumulation of carbon and nitrogen by present-day soil autotrophs.

Evidence for past and present transport of sediment by wind is widespread in the McMurdo Dry Valleys. Geomorphic evidence includes sand dunes, sand sheets, and ventifacts. Inorganic and organic sediment may also be redistributed to lakes and soils by the wind.

Orbital radar data have been used to estimate the potential for sediment transport by wind in other desert regions. The flux of sediment that can be transported by the wind is determined in part by the roughness of the surface. In a similar way, the radar backscatter of unvegetated, dry surfaces is also controlled by surface roughness. Studies of arid terrains in the southwestern United States have shown that the potential for transport of sediments by wind can be estimated by using the correlations between radar backscatter coefficients, surface roughness, and aerodynamic roughness. Data from the Radarsat instrument will be used to estimate the potential for sediment transport by wind in the McMurdo Dry Valleys.

Field measurements of surface roughness characteristics and boundary-layer winds were conducted at 12 sites in the Dry Valleys in December - January 1996-7. Initial analyses of data indicate an excellent relation (r2 = 0.74) between the aerodynamic roughness parameter (zo) and the roughness concentration (lamda). When calibrated Radarsat data are available, we will correlate surface roughness characteristics with radar backscatter coefficients and use these data to produce a map of aeolian sediment transport potential.

Some estimates of actual sediment transport by wind can be derived from the dust pan studies over the past two years. Passive dust collectors installed at Lake Fryxell, Lake Bonney, and Lake Hoare meteorological stations in February 1994 and retrieved in January this year indicate dust deposition rates of between 0.25 and 1 g m-2 yr-1. Sand deposition rates over the same period are approximately 150 g m-2 yr-1. Spectrofluormetric characterization of the organic component of the dust fraction indicates that it is derived from microbial and algal sources. Wind transported materials may therefore be an important source of organic matter for Dry Valleys lakes.

Studies planned for the 1997-98 field season include measurements of winds and surface roughness to test model predictions and installation of a network of dust collectors throughout the Taylor Valley.

Scottnema lindsayae is a soil nematode endemic to the Antarctic cold desert that enters anhydrobiosis when exposed to environmental stress. Anhydrobiosis is an ametabolic survival strategy also used by hot desert nematode species and is characterized by loss of body water and coiled morphology. In field experiments in Taylor Valley, Antarctica, we examined the relationship between anhydrobiosis and soil salinity and moisture, factors that vary greatly in the Dry Valleys as a result of geographic and climatic factors affecting water availability. For three sites sampled in austral summer, 1996 the proportion of nematodes in anhydrobiosis increased as soil moisture and salinity declined. Similarly, the abundance of S.lindsayae in these soils was related to moisture and salinity. Knowledge of the conditions inducing anhydrobiosis is important in determining periods of activity during the austral summer when nematodes are involved in soil biogeochemical processes.

The extreme environment of the ice-free Antarctic Dry Valleys defines the limits to life in soil. In the dry valleys, soil biodiversity is less than in any other terrestrial ecosystem. A conceptual model is presented that defines the soil and environmental conditions determining suitable and unsuitable soil habitats for biota. The majority of soils sampled across the McMurdo Dry Valleys (55%) support soil invertebrates (tardigrades, rotifers, nematodes). Most soils contain only one invertebrate taxa and 2 and 3 taxa communities are rare. Nematodes are more abundant and more widely distributed than tardigrades and rotifers. The species diversity of nematodes is also limited, with only one species, Scottnema lindsayae, a microbial feeder, occurring across all four valleys. Our data indicates that soil conditions may be outside the tolerances of dispersing organisms preventing community establishment, thus creating the patchy distribution of soil biota that uniquely defines the Dry Valley landscape. We suggest that the relationships between species diversity, ecosystem function and the soil environment should be more apparent in the dry valleys of Antarctica than in other terrestrial ecosystems.

MacDonald, L.H., P. Conovitz, and D.M. McKnight, Department of Earth Resources, Colorado State University, Fort Collins, CO (e-mail: pcono@meeker.CNR.ColoState.edu and leemac@cnr.colostate.edu), INSTAAR, University of Colorado, Boulder, CO (e-mail: mcknight@snobear.colorado.edu)

Field observations suggest that the hyporheic zone is an important control on the volume and timing of runoff as well as the development of algal communities. However, there has been no systematic measurement of the seasonal development of the hyporheic zone in the dry valley streams. The purpose of this paper is to: (1) present the disparate evidence for a dynamic hyporheic zone; (2) propose a series of hypotheses to be tested in the 1997-98 field season; and (3) outline our proposed methodology for evaluating the role of the hyporheic zone and testing our hypotheses.

Key evidence for the importance of the hyporheic zone include: observed algal mats adjacent to the stream channel (e.g., McKnight and Tate, 1996); the measured loss of water in the Onyx River during the first part of the 1993-94 flow season (Howard-Williams et al., 1997); the analysis of a tracer experiment in Huey Creek (McKnight and Andrews, 1993); observed recession curves (Conovitz et al., in press); and that water diverted into an abandoned channel did not reach Lake Fryxell until the following year. One can also argue for an expanding saturated zone due to the transfer of sensible heat from the streams to the underlying permafrost.

Our proposed fieldwork will attempt to document the seasonal and spatial variability in the hyporheic zone in selected dry valley streams during the 1997-98 field season. Water level measurements at different cross-sections will provide a direct assessment of the saturated zone, while temperature and electrical conductivity data will be used to assess the connectivity between the stream and adjacent saturated areas. We hope to assess the vertical expansion of the hyporheic zone by repeatedly driving a metal rod through the unfrozen layer to the permafrost surface. Conditions necessary for the growth of algal mats will be assessed by simultaneous measurements of the saturated zone and gravimetric soil moisture. One or more tracer experiments will be conducted and analyzed with a new transient routing model (Runkel et al., in preparation). Suggestions for additional work or alternative methodologies are encouraged.