Andrew G Fountain W Berry Lyons

Introduction

The view of climate change during the Pleistocene and the Holocene was very much different a mere decade ago. With the collection and detailed analyses of ice core records from both Greenland and Antarctica in the early and mid-1990s, respectively, the collective view of climate variability during this time period has changed dramatically. During the Pleistocene, at least as far back as 450,000 years b.p., abrupt and severe temperature fluctuations were a regular occurrence rather than the exception (Mayewski et al. 1996, 1998; Petit et al. 1999). During the Pleistocene, these rapid and large climatic fluctuations, initially identified in the ice core records, have been verified in both marine and lacustrine sediments as well (Bond et al. 1993; Grimm et al. 1993), suggesting large-scale (hemispheric to global) climate restructuring over very short periods of time (Mayewski et al. 1997). Similar types of climatic fluctuations, but with smaller amplitudes, have also occurred during the Holocene (O'Brien et al. 1995; Bond et al. 1997; Arz et al. 2001). What were the biological responses to these changes in temperature, precipitation, and atmospheric chemistry? We must answer this question if we are to understand the century- to millennial-scale influence of climate on the structure and function of ecosystems.

Because the polar regions are thought to be amplifiers of global climate change, these regions are ideal for investigating the response of ecological systems to, what in temperate regions might be considered, small-scale climatic variation. Our knowledge of past climatic variations in Antarctica comes from different types of proxy records, including ice core, geologic, and marine (Lyons et al. 1997). It is clear, however, that coastal Antarctica may respond to oceanic, atmospheric, and

East Antarctic Ice Sheet , -

Figure 16.1 Landsat image of the McMurdo Dry Valleys. Both the east Antarctic Ice sheet in the bottom left of the photo and the Ross Sea ice in the top right of the photo are covered by snow. The darker shades identify the bare soil in the Dry Valleys.

ice sheet-based climatic "drivers," and therefore ice-free regions, such as the McMurdo Dry Valleys, may respond to climate change in a much more complex manner than previously thought (R. Poreda, unpubl. data 2001). Since the initiation of the McMurdo Dry Valleys Long-Term Ecological Research program (MCM) in 1993, there has been a keen interest not only in the dynamics of the present day ecosystem, but also in the legacies produced via past climatic variation on the ecosystem. In this chapter we examine the current structure and function of the dry valleys ecosystem from the perspective of our work centered in Taylor Valley. From this understanding we examine the changes in the ecosystem in response to climatic changes for the past 27,000 years and highlight the importance of past climatic conditions on current ecosystem functioning.

Site Description

The McMurdo Dry Valleys Long-Term Ecological Research (MCM) site is located on the edge of the Antarctic continent (77.5° S, 163° E) in a much larger region known as the Southern Victoria Land (figure 16.1). Only 2% of the Antarctic is ice free (Drewery et al. 1982) and the McMurdo Dry Valleys is the largest ice-free region on the continent with about 2000 km2 of snow-free area (Chinn 1988). The valleys owe their existence to the Transantarctic Mountains that block the ice flow from the East Antarctic Ice Sheet (figure 16.1). A few lobes of the ice sheet penetrate the mountains and reach the valleys. In addition, numerous small alpine gla-

Figure 16.2 Western Taylor Valley looking west. Lake Bonney is at the lower left. Note the ice-free margin (moat) at the lower end of the lake. Central in the background is the Hughes Glacier, flowing off the Kukri Hills. To the extreme left is the Sollas Glacier, and the shadow on the right of the glacier is a cinder cone, which was used by Wilch et al. (1993) to date the glacier activity. (Photograph by Thomas Nylen)

ciers form in the mountains and flow to the valley floor. The valleys are characterized by a rocky-sandy soil devoid of vascular vegetation, thus yielding a stark landscape (figure 16.2). Perennially ice-covered lakes, which are fed from ephemeral streams originating as glacial meltwater, are present in nearly all the valleys (Fountain et al. 1999). The polar climate of the region experiences continual darkness in midwinter and continual sunlight in midsummer. Air temperatures average about -17°C, with the winter minimum about -40°C and the summer maximum a few degrees above freezing (Clow et al. 1988).

Precipitation occurs as snow and can fall at any time of the year. Annual values of snowfall are about 10 cm water equivalent (Keys 1980), most of which is lost to sublimation (Chinn 1993). The biology of the dry valleys exhibits spectacularly low biodiversity and short food chains (Priscu et al. 1999; Virginia and Wall 1999). The soil communities are limited to a few phyla, including rotifers, tardigrades, nema-todes, protozoans, fungi, and bacteria (Freckman and Virginia 1998). Nematodes represent the largest predator in the valleys. The streams support communities of cyanobacteria, eukaryotic algae, and mosses (McKnight et al. 1999). The lakes host only microorganisms and benthic microbial mats (Wharton et al. 1993, Priscu et al. 1999). Small microbial communities inhabit natural melt holes (cryoconite holes) in the glaciers (Wharton et al. 1985) and in the lake ice (Priscu et al. 1998). Although

Figure 16.3 Map of Taylor Valley

life exists at higher latitudes in Antarctica, the dry valleys host one of the last functioning terrestrial ecosystems where streams, lakes, and soils are interconnected both physically and biologically.

The MCM is primarily focused on Taylor Valley, an east- to west-trending valley about 34 km long and 12 km wide (figure 16.3). To the east, Taylor Valley is open to McMurdo Sound, and to the west it is blocked by Taylor Glacier, which flows from the East Antarctic Ice Sheet (figure 16.1). Elevations range from sea level to 2000 m in the mountains that form the northern and southern boundaries of the valley. Elevations of the valley floor range from sea level to about 60 m. The MCM was initiated in 1993 and research compilations can be found in Priscu (1998) and in Bioscience (1999). To understand how climatic change affects the Taylor Valley ecosystem, a more complete physical description of the valley, ecosystem, and their interaction is required.

Physical Setting

Structurally, the dry valley region has acted as a rigid block since the early Tertiary, 50 million years ago (Fitzgerald et al. 1986). Geotectonic uplift in Taylor Valley (Wilch et al. 1993), which is important to ecological processes in the valley, has been relatively slow for the past 2.57 million years. Throughout the Pliocene, small cinder cones erupted in the intermediate regions of middle Taylor Valley and apparently produced little or no ash deposits. Because the cones are located high on the valley sides, generally above 300 m in elevation, and well away from the valley floor, their role in valley ecology is probably small and is not considered here. Dry valley uplift has received close attention because of the possible ramifications for changing regional climate (Behrendt and Cooper 1991). More recent examination by Wilch et al. (1993) argues that, based on their textural characteristics, the cinder cones could not have erupted under water. Therefore, the age and elevation of the cones provide limiting values for the uplift. If uplift occurred at all over the past 2.6 million years, it could not have exceeded 300 m.

A geomorphological model divides the dry valleys into coastal, intermediate, and interior regions (Marchant and Denton 1996). These divisions roughly correspond, at least since mid-Pliocene times (Wilch et al. 1993), to low, intermediate, and high elevations. The low-elevation coastal areas exhibit modern soil movement (e.g., solifluction, ice wedge polygonal patterning), and the depth to ice-cemented sediment is less than 50 cm. The landforms and soils are less than 12,000 years old because they are formed from late-Wisconsin glacial deposits (addressed next). Midelevation intermediate zones exhibit comparatively low soil moisture. Glacier meltwater is infrequent in this zone, and consequently streamflow is uncommon except during rare extremes of warm air temperatures. Water-induced slope movement (solifluction) is limited to areas with a potential water source such as near glaciers or annually forming snow patches. The landscape of intermediate zones is not dynamic, and climatic conditions favor preservation of desert pavements and sand-wedge polygons. Based on potassium-argon dating of in situ ash deposits, slope movement has been minimal in central Taylor Valley during the past 7.1 million years (Marchant and Denton 1996). The high-elevation interior zone, found above 800 m, has virtually no soil moisture, and meltwater is entirely absent. There is no active layer over permafrost because soils lack ice. Dated ash deposits, trapped and buried in thermal contraction cracks in the soil or preserved under desert pavements, exhibit ages of up to 10-15 million years (Marchant and Denton 1996). The sedimentary structure of the ash layer indicates no slope movement or contact with water since the time of deposition. Taylor Valley is completely encompassed by zones 1 and 2 because of its relative proximity to the coast compared to other dry valleys. Nonetheless, the age of the landscape surfaces is quite old compared to its temperate counterparts: It ranges in age from 12,000 years at the valley bottom to 7.1 million years on the upper valley walls.

Since the start of the Pleistocene, significant surface modification in the lower elevations (zone 1) of Taylor Valley, like all of the dry valleys, has been by glacial activity, including water runoff from glaciers. The soils of zone 1, the valley lowlands, formed from glacial deposits or by events directly resulting from glacier activity. Rates of eolian modification seem to be small. Marchant and Denton (1996) date near-surface ash deposits to 10-15 million years in the windy high-elevation zones, indicating that desert pavements greatly inhibit significant wind erosion or deposition. The soils of zone 1 contain marine and lacustrine organic matter at the lower and intermediate levels, and endolithic sources dominate the organic matter in the higher elevations (Burkins et al. 2000).

Establishing the historic context of landscapes is crucial to understanding all ecosystems (Swanson et al. 1988), but this is particularly the case in the MCM

where past climatic variations truly dictate current ecosystem status. Because of its polar location, the primary disturbances in the MCM ecosystem have been climatic, and the landscape pattern has been primarily dictated by climatic, not biotic, processes.

Climate

The dry valleys are considered a polar desert. As in much of Antarctica, precipitation is very low. Measurements in Wright Valley, adjacent to Taylor Valley, show that the annual snowfall ranged from 0.6 to 10 cm water equivalent (Bromley 1985). Precipitation is greatest nearest the coast and decreases inland (Keys 1980; Fountain et al. 1999). Based on snow depths measured along the floor of Taylor Valley after a summer snowstorm and on snow depths measured on Taylor Valley glaciers at 200-300 m in elevation (Fountain et al. 1999), precipitation accumulation decreased at a gradient of -0.06 cm km-1 (water equivalent) with distance from the coast. Typically, snow sublimates before melting, making little contribution to the hydrology. Only in places where snow accumulates to significant depths and is protected from the winds, such as swales, stream channels, or along glacier margins, does it make a contribution, if transient, to the hydrology of the valley.

In Taylor Valley, temperatures range from a few degrees above freezing in late December or January to about -45°C in winter (Clow et al. 1988, Doran et al. 2002b). Average annual temperature in the valley ranges from -16°C to -20°C. One of the important factors that control temperature is the katabatic winds that flow off the East Antarctic Ice Sheet. These foehn-type winds adiabatically warm as the flow from higher elevations and can raise air temperatures by 10 degrees within 15-20 minutes. Thus, winter temperatures are also partly controlled by the frequency of katabatic winds (Doran et al. 2002b). In addition to warming, these winds dramatically reduce the humidity and significantly increase sublimation from the ice surfaces (Clow et al. 1988). Aside from katabatic effects, the valleys are typically windy. Monthly average wind speeds range from 2 to 4 m s-1 in Taylor Valley (Clow et al. 1988).

Glaciers

Almost all of the glaciers in the dry valleys are relatively small alpine glaciers, a few km2 in area. In Taylor Valley, the glaciers originate from the Asgard Range on the north side and from the Kukri Hills on the south side (figure 16.1), where snow accumulation exceeds loss by sublimation and wind erosion. Taylor Glacier is the largest glacier in the valley, defining its western boundary, and flows from the East Antarctic Ice Sheet. Roughly one-third of Taylor Valley, as defined from the ridge divide of each mountain range, is ice covered (Fountain et al. 1998). The glaciers are polar glaciers with bases frozen to substrate. The larger glaciers terminate in vertical ice cliffs about 20 m high. Unlike the ice of temperate glaciers, the ice is remarkably clean and free of debris. Because the ice is well below freezing point, meltwater is restricted to the glacier surface and flows off the glacier edge. These characteristics contrast with alpine glaciers in the temperate latitudes, which are at their melting point throughout and are not frozen to their bed. Instead, they slide over the substrate. Meltwater enters the body of temperate glaciers, flows internally, and finally emerges from subglacial tunnels at the glacier edge (Fountain and Walder 1998).

The mass exchange of these polar glaciers is relatively small, compared to temperate glaciers. Our observations indicate that about 10 to 30 cm of snow accumulates in the upper zones and about 6 to 15 cm is lost from the ablation zone. These values are consistent with results from previous studies in the adjacent Wright Valley (Bull and Carnein 1970; Chinn 1980). The snow in the upper reaches of the glaciers is cold and dry, and no snowmelt has been observed directly. Mass loss in the upper snow zone is by sublimation because no melting occurs. In the lower elevation (ice-exposed) zone of the glacier, the mass loss is dominated by sublimation and melting. Results from the Canada Glacier indicate that during the summer, sublimation accounts for 40-80% of the ice mass loss, the remainder being lost to melting (Lewis et al. 1998). Since 1993, meltwater production has been limited to the lower fringe of the glaciers within a few hundred meters of the ice edge. Certainly, no melt-water has been observed on the glaciers at elevations above 250-500 m.

Hydrology

Meltwater flows off the glacier to form streams at the base of the ice cliffs. These streams are ephemeral and typically flow for 4-10 weeks a year (McKnight et al. 1999). They transport water, sediment, and nutrients to terminal lakes (figure 16.4). As previously mentioned, snowfall on the valley floor does not contribute significantly to the streams because it usually sublimates before melting (Chinn 1980). However, winter drifts of snow piled against the glacier termini or in stream channels contribute to streamflow in early spring before disappearing by early summer (Fountain et al. 1998). The streams are channelized and flow over continuous permafrost, which occurs at shallow depths of a few tens of centimeters (McKnight et al. 1999). Therefore, groundwater flow is probably limited to the near-surface hy-porheic zones, the saturated zone adjacent to and under the stream channel. Because of the shallow depth and lack of lateral groundwater inflow, the hyporheic zone can extend laterally for several meters on either side of the channel.

The glacial meltwater that feeds the streams typically has a very low solute content, on the order of 1-10 micro Molar (|M) per chemical species (Lyons et al. 1998). The solute content of the glaciers is controlled, in part by the chemistry of the snow accumulation but is otherwise dominated by recycled solutes blown onto the glacier surfaces from the valley floor (Lyons et al. in press). Preliminary analysis of the spatial pattern of glacier meltwater chemistry suggests that it is controlled by the pattern of sediment on the glacier (M. Tranter, unpubl. data, 2001). The mass flux of the stream water increases by one to two orders of magnitude before it reaches the lake as a result of salt dissolution and chemical weathering processes in the channel and within the hyporheic zone (Lyons et al. 1998; Gooseff 2002). Evaporative losses in the stream channel also help to increase the concentration of solutes (B. Vaughn, unpubl. data, 1993).

Three main lakes occupy the bottom of Taylor Valley (lakes Fryxell, Hoare, and

Figure 16.4 Map of the streams in Taylor Valley, draining from the glacier (gray) into the lakes (black). McMurdo Sound is the ocean area.

Bonney) in addition to numerous smaller lakes and ponds that dot the landscape. All lakes and ponds have permanent ice cover (3-6 m thick), and the smaller water bodies are probably completely frozen during winter. As previously mentioned, the lakes are terminal, and lose water only through sublimation of the ice surface and by evaporation from the narrow fringe of open water around the perimeter of each lake in late summer. As with terminal lakes elsewhere (e.g., Great Salt Lake), these lakes are sensitive to small changes in water inflow. Since the beginning of measurement recording in the 1970s by Chinn (1993), generally all lakes in the dry valleys have been rising. Only recently have the lakes slowed or stopped rising (figure 16.5). For Lake Hoare, the level has dropped over the past few years in response to cooler summer weather and increased summer snowfall (Fountain et al. 1999; Doran et al. 2002a). Because the perennial ice effectively covers the lake water, there are no wind-generated currents (Hawes 1983). The exchange of gases between the lake water and the atmosphere is therefore restricted (Wharton et al. 1986), and light penetration into the lake water is reduced (Howard-Williams et al. 1998). The major lakes in Taylor Valley are density stratified because of salinity gradients (Spigel and Priscu 1998). Some mixing occurs in summer at the lake edge, where the ice melts and localized turbulence can propagate into the lake (Miller and Aiken 1996). The motion of the lake water is limited to slow horizontal movement. The melted lake fringe also allows a limited exchange of gases with the atmosphere. In addition, streamflow enters the lake through the melted fringe, supplying the lake with nutrients and water (Tyler et al. 1998).

as OL

1970

1975

1980

1985 Year

1990

1995 2000

Figure 16.5 Lake level change in Taylor Valley.

1975

1980

1985 Year

1990

1995 2000

Figure 16.5 Lake level change in Taylor Valley.

Soils

A pebble to boulder (1 to 20 cm diameter) surface pavement covers a coarse, but variable, soil texture with little cohesion and little organic content (Bockheim 1997; Campbell et al. 1998). Permafrost is ubiquitous in the valleys, and in soils it is typically found under a 10- to 30-cm-thick active layer. In Taylor Valley, the soil texture is dominated 95-99% by coarse-grained sand (Burkins et al. 2000). The high variability of the soil texture results from glacial action in the valley, as described subsequently. The organic content of the soils does not exceed 0.03% by weight of organic carbon content (Burkins et al. 2000). The moisture content of the soil is low and varies with respect to the distance from the coast. In the coastal region, soil moisture averages about 1% by weight; farther inland on the valley floor soil moisture drops to 0.5% (Campbell et al. 1998). Smaller scale patterns of soil moisture depend on proximity to water sources such as lakes and ephemeral streams. In addition, winter snow accumulations, which are typically thin and quite patchy, provide a transient water source. However, the effect of these sources does not extend beyond 10 m (Campbell et al. 1998; McKnight et al. 1999), and 95% of the soil does not receive liquid water (Campbell et al. 1998). Moisture for these soils depends on either brief accumulations of snow, which typically sublimate before melting, or sublimation of ice-cemented permafrost at depth (Campbell et al. 1998; McKay et al. 1998).

Ecosystems

Ecosystems in each of the dry valleys is relatively isolated because high ice-capped mountains separate the valleys and water drains internally within each valley. The lack of transport between valleys is suggested by mitochondrial DNA analysis of the soil nematode, Scottnema lindsayae. It was thought that the genetics of this nematode would be fairly uniform because it is easily dispersed by winds and streams, but soil samples from five different valleys exhibited significant mito-chondrial differences, suggesting that the gene flow among the populations is restricted (Courtright et al. 2000). Within valleys, linkages among soil, stream, and lake ecosystems are relatively weak compared to valleys in temperate regions. In Taylor Valley, overland flow and groundwater are nonexistent, therefore the soils are connected to the streams or lakes only through the hyporheic zone interactions during the few weeks of streamflow each year.

One integrating process, although weak, is the wind. Eolian processes transport sediment and organic material across the valley. The wind entrains organisms (Virginia and Wall 1999) from the soil and erodes algal mats and mosses from the stream channels (McKnight et al. 1999). Benthic algal mats in the lakes rise from the lake bottom as a result of the buoyancy caused by accumulated gases in the mats (Parker et al. 1982) and become frozen in the lake ice at the surface. Through continual freezing on the bottom and sublimation at the top, these mats are transported through the ice and become exposed at the surface where they are wind eroded. All these materials are transported across the valley and are redeposited in soils, stream channels, and lakes. It is common to find bits of algal mat on Commonwealth Glacier at 300 m in altitude. We hypothesize a net transport of sediment and organic material down valley toward the coast due to the high velocity wind events caused by katabatic winds flowing off the ice sheet. Although a down-valley gradient of increasing organic carbon is observed (Virginia and Wall 1999), local modifications, as explained by climatic events (which we discuss later in this chapter), also can contribute to this trend.

The polar climate of the dry valleys includes a long winter of darkness, very cold temperatures, and no surface water. During this period the linkages between the ecosystems are severed. Certainly winds continue to disperse organic matter, but the ecosystems are no longer integrated and the only functioning ecosystem, the lakes, is completely isolated. Each system adopts a strategy to survive the winter in isolation. Therefore, in such an environment the biological linkages between systems must necessarily be weak because they are severed seasonally. The soil ne-matodes enter a state of anhydrobiosis, which allows them to survive a winter of no water and subfreezing temperatures (Crowe 1971). They have been known to survive in this state for over 60 years (Freckman 1986). Stream mosses and algal mats also become freeze-dried after the meltwater supply ceases and the streams stop flowing. After rewetting, some mats began to photosynthesize in 10-20 minutes (Vincent and Howard-Williams 1986; Hawes et al. 1992). Recent experiments show that stream algal mats that have not experienced water for 25 years reactivated a full microbial ecosystem over a 1.5-km stream reach within 2 weeks of rewetting (McKnight et al. 1999). In winter, the lakes become isolated from the external environment as the fringe of open water around the lakes and within the perennial lake ice freezes. For the rest of the winter, the lake environments are entirely isolated from any exchange with the outside. Moreover, little light exists in the polar autumn and spring, whereas complete darkness occurs in the middle of winter. Phy-

toplankton in the lakes, which depend on light for photosynthesis, turn to alternative sources of energy. Studies on Antarctic lakes, particularly those in the dry valleys, show that some phytoplankton species will ingest bacteria, a source of organic carbon, during low light conditions (Roberts and Laybourn-Parry 1999; Laybourn-Parry et al. 2000). This mixotrophic strategy allows the phytoplankton to endure the winter darkness. Our understanding of midwinter biologic processes in the lakes is incomplete because of the logistical limitations that to date have precluded over-winter studies. In spring phytoplankton growth is triggered solely by increased solar radiation because vertical mixing is largely absent. Biomass and phytoplankton production generally increase throughout the summer through late January.

Century- to Millennial-Scale Climate Changes

Climate change in the dry valleys has been largely inferred from geomorphic evidence of past glacier positions (e.g., Stuiver et al. 1981; Denton et al. 1989) and lake level heights (e.g., Stuiver et al. 1981; Hall and Denton 1995, 2000). In addition, profiles of chemical concentrations in the lakes have been used to infer past lake drawdowns (Wilson 1964; Lyons et al. 1998). More recently, isotopic results from ice cores (Steig et al. 2000) and temperature measurements in boreholes in the bedrock (G. Clow, unpubl. data, 1993) and in glaciers (Clow and Waddington 1996) have been used to more directly infer climate changes in the region. We summarize the millennial-scale climate changes based on all this evidence, using the ice core data as our baseline chronology.

An ice core was obtained from Taylor Dome, about 140 km from Taylor Valley. Although a number of cores have been obtained from the continent (e.g., Jouzel et al. 1987; Morgan et al. 1997; Mayewski et al. 1996; Legrand and Mayewski 1997), this particular core is closest to the dry valleys and provides better information on regional climatic variations than cores from much farther away. In fact, the results from the Taylor Dome core contrast with results from cores at Vostok and Byrd (Blunier et al. 1997, 1998), which reside on opposite sides of the Antarctic continent away from Taylor Dome. Climatic data inferred from the other cores are out of phase with that from Greenland, whereas data from the Taylor Dome core are in phase. These differences have been attributed to the spatially variable deep ocean currents that transport heat between the southern ocean and the lower latitudes (Steig et al. 1998).

A variety of geochemical analyses have been completed on the Taylor Dome core, but we concentrate here on those results of importance to the climate of the dry valleys as summarized by Steig et al. (2000). Figure 16.6 shows the inferred air temperatures from oxygen isotopes and the snow accumulation record from the ice core. Starting about 60 thousand (kyr) ago (calendar years) a slow cooling began, which culminated in the last glacial maximum. During the same time period, snow accumulation decreased, reaching a minimum during the last glacial maximum. Rapid warming, known as the B0lling-Aller0d event, occurred about 15 kyrs ago at the termination of the glacial period. Subsequently, the Younger-Dryas cooling

Figure 16.6 Top: Accumulation rate in cm of ice per year (cm ice a-1) at Taylor Dome based on 10Be data. Bottom: Oxygen isotope values (S180) from Taylor Dome (left axis) and inferred temperature change (AT) (right axis). The timescale is in thousands of years before present. Note the shift in timescale at 20,000 years. These data are from Steig et al. 2000, figure 7 and figure 6, respectively.

Figure 16.6 Top: Accumulation rate in cm of ice per year (cm ice a-1) at Taylor Dome based on 10Be data. Bottom: Oxygen isotope values (S180) from Taylor Dome (left axis) and inferred temperature change (AT) (right axis). The timescale is in thousands of years before present. Note the shift in timescale at 20,000 years. These data are from Steig et al. 2000, figure 7 and figure 6, respectively.

event dropped air temperatures by about 4°C before returning to the warmer B0lling-Aller0d values by about 11 kyr ago. Substantial cooling events occurred at 9.5 kyrs and 6.5 kyrs ago. From the B0lling-Aller0d event through the cooling event of 6.5 kyrs ago, accumulation generally increased. A general cooling trend and decreasing accumulation has dominated the late Holocene. Details of the temperature trend for the late Holocene (past few kyrs) are not available from the Taylor Dome ice core at this time because of potential nontemperature influences in the isotope data (Waddington, pers comm., 2000). Fortunately, subsurface earth temperatures were measured in a borehole in Taylor Valley, and based on temperature variations with depth, a record of surface temperatures was obtained (Clow 1998). A general cooling persisted in the valley from about 4 kyrs ago to about 1 kyr. Since that time the air temperatures have warmed.

These climatic trends have caused changes in the regional ice sheet extent and in lake levels. With the cooling of the last glacial maximum, the Antarctic Ice Sheet began to enlarge and advance. By about 23.8 (C14) kyrs ago (~ 27 kyrs), the Ross Ice Shelf entered Taylor Valley from the ocean (Hall and Denton 2000; Stuiver et al. 1981). The ice shelf blocked the seaward opening of the valley and dammed an inland lake, known as Lake Washburn, to a level of about 200 m deep. The lake existed until about 8.3 (C14) kyrs ago (~9.5 kyrs) when the ice shelf receded from the valley and the lake waters drained from the valley. The valley has cooled 1.2°C since about 4 kyrs ago; this probably led to the major drawdown of the lakes in the dry valleys, which reached their maximum drawdown 0.9-1.2 kyrs ago (Wilson 1964; Lyons et al. 1998). Lake Hoare either completely desiccated or did not exist prior to ~2 kyrs ago (Lyons et al. 1998). Since that time, the lakes have refilled as air temperatures increased about 2°C in the past 1000 years. These recent changes in the valleys are well correlated with other inferred climate changes in the region. For example, the "penguin optimum," associated with rookery abundance and a warmer climate, occurred between 3 and 4 kyrs ago and terminated about 3 kyrs ago (Baroni and Orombelli 1994). The rookeries were reoccupied starting about 1200 years ago.

Humans first visited the dry valleys in 1903 when Robert F. Scott's western journey included a side trip down the Taylor Glacier to Lake Bonney. Members of the party measured the width of the lake at its narrowest point, which was later used by Chinn (1993) to estimate a lake level. Since 1903 the lakes have risen, indicating a generally warmer climate. The trend has continued from the 1970s, when Chinn (1993) started taking measurements, through the 1980s. Thinning of the lake ice suggest a 2°C warming during the 1980s, although other factors may have been important (McKay et al. 1985). This warming may have caused increased melt and runoff, which in turn caused increasing lake levels. Between 1986 and 2000, air temperatures have decreased by 0.7°C per decade, runoff from glaciers has slowed, and the lake levels have generally stopped rising and in some cases are falling (Doran et al. 2002a; Welch et al., chapter 10 this volume).

Ecosystem Response

The formation of Lake Washburn about 27 kyrs ago created a lake that extended from the valley mouth to Taylor Glacier at the western end (figure 16.7). Phyto-plankton in the water column and benthic algal mats apparently inhabited this enlarged lake because Burkins et al. (2000) found that the pattern of lacustrine and marine sources of organic carbon in soil transects correlated with the former lake extent. The total lacustrine biomass during the time of Lake Washburn was much larger than it is at present. The organic carbon from this biomass is still present in the soils as indicated by Burkins's work. Therefore, the current carbon production in the valley is very slow. Heretofore, it was thought that wind-transported organic carbon (lacustrine microbial mats, endolithic communities, and stream-based mats and mosses) in the valley was a major source of carbon transfer. Although real-time observations suggest that this is true, the significance of this process is disputed by the correlation of the lacustrine-derived organic carbon and the extent of Lake Washburn (Burkins et al. 2000). Given the lack of higher plants and animals, soil organisms are apparently sustained by the organic carbon in the soils left by Lake Washburn. Therefore, the dry valley lakes are connected to the soils across time rather than across space.

Once the Ross Ice Shelf retreated from Taylor Valley about 9.5 kyrs ago, the ice no longer blocked the valley, and Lake Washburn drained, leaving smaller lakes

ponded in the low spots along the valley floor. The climate began to cool about 4 kyrs ago and the streamflow from the glaciers slowed below that required to maintain the lakes. Because of this, the lakes began to decrease in size. They reached a minimum by ~1000 years ago. If Lake Hoare existed prior to that time, it had completely evaporated and the residual salts were blown away (Lyons et al. 1998). We presume that the stream ecosystems were largely absent and the soil ecosystem had been significantly reduced by this time. Whether all terrestrial biota entered a long-term anhydrobiosis cannot be ascertained. Lakes Fryxell and Bonney evaporated down to a brine pool, concentrating the salts and nutrients such as organic and inorganic carbon. Since that time, the climate warmed and the greater meltwater fluxes increased in the lake. The fresh meltwater did not mix with the concentrated brine pools, due to density differences, and floated over the top, trapping the brine at depth. Diffusion fluxes from the concentrated brines at depth have been used to obtain the refilling age of ~1 kyr bp (Wilson 1964; Lyons et al. 1998).

Studies of the phytoplankton in Lake Bonney indicate that its productivity and biomass increase when solar radiation penetrates the ice cover early in the season (Priscu et al. 1999). This increase first occurs in the shallow waters just underneath the ice surface. Overall, less than 0.1% of the dissolved carbon pool in the trophogenic zone is contributed by streamflow (Priscu et al. 1999). As light intensity increases with the approach of the austral summer solstice, the main region of productivity shifts from the surface 15 m down the water column to the bottom of the trophogenic zone (figure 16.8). This shift results from sufficient light reaching

Figure 16.8 Water column properties for the east lobe of Lake Bonney. (a) Photosyntheti-cally available radiation. (b) Primary productivity, (c) Chlorophyll a. (From Priscu et al. 1999)

the depth of relatively high flux of inorganic and organic carbon, through molecular diffusion, from the pools of concentrated nutrients below (Priscu 1995; Priscu et al. 1999). We presume that the phytoplankton continue to be productive at this depth into the Antarctic autumn after the streams stop flowing. The depth of productivity may change to more shallow levels again in response to low light levels of the Antarctic winter. Unfortunately, we lack measurements during this period and during the winter because of logistical constraints. In any case, it is clear that the current lake ecosystem is strongly conditioned by past climatic events. The drawdown of the lakes about 1000 years ago concentrated the lake nutrients into pools on which modern phytoplankton now depend.

The effect of past imprints on current ecosystem functioning is known as a legacy (Vogt et al. 1997). In our case, the legacy involves the change in climate and its impact on lake size. Formerly large lakes supply nutrients to lowland soils, and the remnants of paleolakes supply nutrients to the current lakes. The vital importance of climatic legacy in the dry valleys is due to its extreme environment, low biodiversity, and short food chains. The extreme polar environment reduces the amount of water and energy available to the ecosystem, greatly slowing the rate of nutrient cycling. The low biodiversity and short food chains make the ecosystem directly dependent on the physical environment such that few buffers exist and the response of the ecosystem to slight climatic change is immediate. This is illustrated by the current decrease in lake primary productivity as a result of a cooling in climate (Doran et al. 2002a). The cooling trend decreased the magnitude of meltwa-ter produced on the glaciers, which in turn decreased the streamflow and nutrient flux to the lakes. Lake levels have stopped rising and, in some cases, are falling. It should be emphasized that the ice core and borehole temperature data indicate that these massive changes in lake size have been driven by annual changes of 2°C or less. Again, we emphasize that in more temperate settings these are small changes, but in the polar environment these relatively small variations are amplified to bring about very large environmental responses. In the Taylor Valley this is due, in part, to the fact that this ecosystem is so closely dependent on the change of state of water and that, during the austral summer, very small temperature changes can greatly influence the hydrologic system.

General Ecosystem Questions and Summary

At the Workshop on Century- to Millennial-Scale Climate Change and Ecosystem Response, as part of the Long-Term Ecological Research program's All Scientist Meeting in Snowbird, Utah (2-4 August 2000), several general ecosystem questions were asked. These are summarized here for the McMurdo Dry Valleys LTER.

What preexisting conditions affect the significance of the climatic event? Our food web is populated by microbiota responding to microenvironmental variations that change significantly over short distances and over time. The lack of higher plant communities tie our terrestrial ecosystems more closely to geologic structures than in more temperate regions. Also, a nutrient-poor environment with weak spatial linkages and slow geomorphic change enhances the linkages over time (legacy) (Morehead et al. 1999).

Are the effects direct or cascading? Our current understanding is that the effects are direct because of the short food chains. Poor linkages that would otherwise buffer the direct effects are absent.

Is the ecosystem response completed before the next event? Events of different magnitudes occur over different timescales, each with their own response time. However, the current legacy of the Lake Washburn event (27,000 years ago) is contemporaneous with the current legacy of the lake drawdown 1200 years ago. Given the extremely slow cycling of nutrients and the pace of geomorphic change, we suspect that ecosystem responses are overprinted on each other and are not completed before the next event occurs.

Does the system return to its original state after the event? We don't actually know, but the answer depends on whether we are currently in a state of perturbed static equilibrium, whereby a constant state exists to which the system can return, or whether the system is in a state of dynamic equilibrium, where there is no constant state but the system varies about some constant value. Given the relatively short record of direct observations, the patchy record of past events, and the slow rate of change in the valleys, it is impossible to tell at this time. However, we can say that the current system is perturbed, given the important legacies recently revealed.

Conclusions

The ecosystem of the McMurdo Dry Valleys exhibits a strong and clear dependence on past climatic conditions. Relatively large variations in climate have concentrated nutrients that cannot be attained under "normal" or static conditions. Because of the nutrient-poor, energy-limited environment of the dry valleys, past concentrations of nutrients play a major role in current ecosystem structure and function. This finding draws attention to the need for understanding past climatic conditions, at least in extreme polar environments, to interpret current ecosystems. The ecosystems within the dry valleys are certainly poorly linked, and the soils and lakes appear to behave completely independently. Only through our understanding of legacy do we realize that the systems are linked across time in a very real and vital manner.

In addition, we observe a polar amplification whereby small changes in climate produce large variations in ecosystem response. The energy limitation in the dry valleys keeps much of the potential water supply locked in the frozen reservoirs of the surrounding glaciers. Small changes in the summer surface energy balance of the glaciers, resulting in the surface temperature rising above the melting point, create a large water flux. Conversely, a small change that results in lowering the temperature below the melting point significantly reduces water production. The summer temperatures of the dry valleys hover around the melting point, and this highly nonlinear response to changes in temperature is a normal condition of the valleys. Although our examination of past climatic effects on ecosystem response does not explicitly address this process, we know from our field experiences that this nonlinear response is at the heart of all observed changes. The creation of Lake Washburn must have occurred when local energy levels allowed the ice temperature to warm and sustain melting. Conversely, the energy levels during lake drawdown must have dropped sufficiently to reduce the ice temperatures below the freezing point for much of the summer periods.

What do these research findings have to offer toward understanding ecosystems in more energy- and water-rich environments? Legacies, including climatic, anthropogenic, and volcanic events, are probably important at all spatial and temporal scales. The significance of each legacy is dependent on the magnitude of the change and response time of the landscape and ecosystem to accommodate that change and return the entire system back to preexisting conditions. Perhaps by understanding the impact and residence time of natural changes, we can anticipate the effect of anthropogenic change. In addition, our research may help to frame how life might survive extreme events such as a bollide impact or a "snowball" earth condition. When ecological linkages are severed, some resilience is provided by legacy effects that store nutrients that would otherwise not be available.

Acknowledgments This chapter would not have been possible without the insight and labors of our LTER colleagues and graduate students. This chapter is dedicated to them. We also appreciate the superb effort of Antarctic Support Associates, who provided much of the logistical support that made this project possible. Funding for this project was provided by the National Science Foundation, Office of Polar Programs, grant McMurdo Dry Valleys: A Cold Desert Ecosystem, OPP 9211773, OPP 9813061; and The Role of Natural Legacy on Ecosystem Structure and Function in a Polar Desert: The McMurdo Dry Valleys LTER Program OPP 9810219, OPP 0096250.

References

Arz, H. W., S. Gerhardt, J. Patzold, and U. Rohl. 2001. Millennial-scale changes of surface and deep water flow in the western tropical Atlantic linked to Northern Hemisphere high-latitude climate during the Holocene. Geology 29:239-242.

Baroni, C., and G. Orombelli. 1994. Abandoned penguin rookeries as Holocene paleocli-matic indicators in Antarctica. Geology 22:23 -26.

Behrendt, J. C., and A. Cooper. 1991. Evidence of rapid Cenozoic uplift of the shoulder escarpment of the Cenozoic West Antarctic rift system and a speculation on possible climatic forcing. Geology 19:315-319.

BioScience. 1999. (Special section on the McMurdo Dry Valleys). BiosScience 49:9591019.

Blunier, T., and 9 others, 1997. Timing of the Antarctic cold reversal and atmospheric CO2 increase with respect to the Younger Dryas event. Geophysical Research Letters 24:2683-2686.

Blunier, T., and 10 others, 1998. Asynchrony of Antarctic and Greenland climate change during the last glacial period. Nature 394:739-743.

Bond, G., and 13 others, 1993. Evidence for massive discharges of icebergs into the North Atlantic Ocean during the last glacial period. Nature 360:245-249.

Bockheim, J. G. 1997. Properties and classification of cold desert soils from Antarctica. Journal of the Soil Science Society of America 61:224-231.

Bond, G., W. Showers, M. Cheseby, R. Lotti, P. Almasi, P. de Menocal, P. Priore, H. Cullen, I. Hajdas, and G. Bonani. 1997. A pervasive millennial-scale cycle in North Atlantic Holocene and glacial climates. Science 278:1257-1266.

Bromley, A. A. 1985. Weather observations: Wright valley, Antarctica. Informational Publication 11. New Zealand Meteorological Service. Wellington, New Zealand.

Bull, C., and C. R. Carnein. 1970. The mass balance of a cold glacier: Meserve Glacier, south Victoria Land, Antarctica. International Association of Hydrological Sciences 86:429446.

Campbell, I. B., G. G. C. Claridge, D. I. Campbell, and M. R. Balks. 1998. The soil environ ment of the McMurdo Dry Valleys, Antarctica. In Priscu, J. C. (editor) Ecosystem Dynamics in a Polar Desert. American Geophysical Union. Antarctic Research Series 2:297-322.

Burkins, M. B., R. A. Virginia, C. P. Chamberlain, D. H. Wall. 2000. The origin of soil organic matter in Taylor Valley, Antarctica. Ecology 81:2377-2391. Chinn, T. J. 1980. Glacier balances in the dry valleys area, Victoria Land, Antarctica. International Association of Hydrological Sciences 126: 237-247. Chinn, T. J. 1988. The "Dry Valleys" of Victoria Land, in R. S. Williams, and J. G. Ferrigno. Satellite image atlas of glaciers of the world, Antarctica. U.S. Geological Survey Professional Paper 1386-B:B39-B41. Chinn T. J. 1993. Physical hydrology of the Dry Valley lakes. Pages 1-52 in W. J. Green and E. I. Friedmann, editors. Physical and Biological Processes in Antarctic lakes. Antarctic Research Series 59. American Geophysical Union, Washington, DC. Clow, G. D. 1998. Reconstructing past climatic changes in Greenland and Antarctica from borehole temperature measurements [abs]. EOS—Transactions of the American Geophysical Union 79:F833. Clow, G. D., C. P. McKay, G. M. Simmons, Jr., and R. A. Wharton, Jr. 1988. Climatological observations and predicted sublimation rates at Lake Hoare, Antarctica. Journal of Climatology 1:715-728.

Clow, G. D., and E. D. Waddington. 1996. Acquisition of borehole temperature measurements from Taylor Dome and dry valleys for paleoclimate reconstruction. Antarctic Journal of the United States 31:71-72. Courtright, E. M., D. H. Wall, R. A. Virginia, L. M. Frisse, J. T. Vida, and W. K. Thomas. 2000. Nuclear and Mitochondrial DNA Sequence Diversity in the Antarctic Nematode Scottnema lindsayae. Journal of Nematology 32:143-153. Crowe, J. H., 1971. Anhydrobiosis: An unsolved problem. American Naturalist 105:563571.

Denton, G. H., J. G. Bockheim, S. C. Wilson, and M. Stuiver. 1989. Late Wisconsin and early Holocene glacial history, inner Ross Embayment, Antarctica. Quaternary Research 31:151-182.

Doran, P. T., J. C. Priscu, W. B. Lyons, J. E. Walsh, A. G. Fountain, D. M. McKnight, D. K. Moorhead, R. A. Virginia, D. H. Wall, G. D. Clow, H. G. Fritsen, C. P. McKay, A. P. Parsons. 2002a. Antarctic climate cooling and terrestrial ecosystem response. Nature 415:517-520.

Doran, P. T., C. P. McKay, G. D. Clow, G. L. Dana, A. G. Fountain, T. Nylen, W. B. Lyons. 2002b. Climate observations from the McMurdo Dry Valleys, Antarctica, 1986-2000. Journal of Geophysical Research-Atmospheres. 107:1-12. Drewery, D. J., S. R. Jordan, and E. Jankowski. 1982. Measured properties of the Antarctic ice sheet: Surface configuration, ice thickness, volume and bedrock characteristics. Annals of Glaciology 2:834-891. Fitzgerald, P. G., M. Sandiford, P. J. Barrett, and A. J. W. Gleadow. 1986. Asymmetric extension associated with uplift in the Transantarctic Mountains and Ross Embayment. Earth and Planetary Science Letters 81:67-78. Fountain, A. G, G. L. Dana, K. J. Lewis, B. L. Vaughn, and D. M. McKnight. 1998. Glaciers of the McMurdo Dry Valleys, Southern Victoria Land, Antarctica. Pages 65-76 in J.

C. Priscu, editor. Ecosystem Dynamics in a Polar Desert: The McMurdo Dry Valleys, Antarctica. Antarctic Research Series 72, American Geophysical Union, Washington,

Fountain, A. G., and 12 others. 1999. Physical controls on the Taylor Valley ecosystem, Antarctica. BioScience 49:961-971.

Fountain, A. G., and J. S. Walder. 1998. Water flow through temperate glaciers. Reviews of Geophysics 36:299-328.

Freckman, D. 1986. The ecology of dehydration in soil organisms. Pages 157-168 in A. C. Deopold, editor. Membranes, metabolism, and dry organisms. Cornell University Press, Ithaca, New York.

Freckman, D., and R. Virginia. 1998. Soil biodiversity and community structure in the McMurdo Dry Valleys, Antarctica. Pages 323-336 in J. C. Priscu, editor. Ecosystem Dynamics in a Polar Desert: The McMurdo Dry Valleys, Antarctica. Antarctic Research Series 72, American Geophysical Union, Washington, D.C.

Gooseff, M. N., W. B. Lyons, A. E. Blum, and D. M. McKnight. 2002. Weathering reactions and hyporheic exchange controlling stream water chemistry in a glacial meltwater stream in the McMurdo Dry Valleys. Water Resources Research, 38:15-1-15-17.

Grimm, E.C., G. L. Jacobson, Jr., W. A. Watts, B. C. S. Hansen, and K. A. Maasch. 1993. A 50,000 record of climate oscillations from Florida and its temporal correlation with the Heinrich events. Science 261:198-200.

Hall, B., and G. H. Denton. 1995. Late Quaternary lake levels in the Dry Valleys, Antarctica. Antarctic Journal of the United States 30:52-53.

Hall, B., and G. H. Denton. 2000. Radiocarbon chronology of Ross Sea Drift, eastern Taylor Valley, Antarctica: Evidence for a grounded ice sheet in the Ross Sea at the last glacial maximum. Geografiska Annaler 82A:305-336.

Hawes, I. 1983. Turbulence and its consequences for phytoplankton development in two ice covered Antarctic lakes. British Antarctic Survey Bulletin 60:69-81.

Hawes, I., C. Howard-Williams, and W. F. Vincent. 1992. Desiccation and recovery of Antarctic cyanobacterial mats. Polar Biology 12:587-594.

Howard-Williams C., A. Schwarz, I. Hawes, and J. C. Priscu. 1998. Optical properties of lakes of the McMurdo Dry Valleys. Pages 189-204 in J. C. Priscu, editor. Ecosystem Dynamics in a Polar Desert: The McMurdo Dry Valleys, Antarctica. Antarctic Research Series 72, American Geophysical Union, Washington, D.C.

Jouzel, J., C. Lorius, J. R. Petit, C. Genthon, N. I. Barkov, V. M. Kotlyakov, and V. M. Petrov. 1987. Vostok ice core: A continuous isotope temperature record over the last climatic cycle (160,000 years). Nature 329:403-407.

Keys, J. R. 1980. Air temperature, wind, precipitation and atmospheric humidity in the McMurdo region. Department of Geology Publication No. 17 (Antarctic Data Series No. 9), Victoria University of Wellington, New Zealand.

Laybourn-Parry, J., E. C. Roberts, and E. M. Bell. 2000. Mixotrophy as a survival strategy among planktonic protozoa in Antarctic lakes. Pages 33-40 in W. Davidson, C. Howard-Williams, and P. Broady, editors. Antarctic ecosystems: Models for wider ecological understanding. Caxton Press, Christchurch, New Zealand.

Legrand, M., and P. A. Mayewski. 1997. Glaciochemistry of polar ice cores: A review. Reviews of Geophysics 35:219-243.

Lewis, K. J., A. G. Fountain, and G. L. Dana. 1998. Energy balance studies of Canada Glacier, Taylor Valley, McMurdo Dry Valleys, Antarctica. Annals of Glaciology 27:603609.

Lyons, W. B., P. A. Mayewski, L. R. Bartek, and P. T. Doran. 1997. Climate history of the McMurdo Dry Valleys since the Last Glacial Maximum. Pages 15-22 in C. HowardWilliams, W. B. Lyons, and I. Hawes, editors. Ecosystem Processes in Antarctic IceFree Landscapes. A.A. Balkema, Rotterdam, Netherlands.

Lyons, W. B., S. W. Tyler, R. A. Wharton, Jr., D. M. McKnight, and B. H. Vaughn. 1998. A late Holocene desiccation of Lake Hoare and Lake Fryxell, McMurdo Dry Valleys, Antarctica. Antarctic Science 10: 247-256.

Lyons, W. B., K. A. Welch, A. G. Fountain, G. D. Dana, B. V. Vaughn, and D. M. McKnight, in press, Surface glaciochemistry of Taylor Valley, Southern Victoria Land, Antarctica, and its relationship to stream chemistry. Hydrological Processes.

Lyons, W. B., K. A. Welch, C. A. Nezat, D. M. McKnight, K. Crick, J. K. Toxey, J. A. Mastrine. 1998. Chemical weathering rates and reactions in the Lake Fryxell Basin, Taylor Valley: Comparison to temperate river basins. Pages 147-154 in C. Howard-Williams, W. B. Lyons, and I. Hawes, editors. Ecosystem Processes in Antarctic Ice-Free Landscapes. A.A. Balkema, Rotterdam, Netherlands.

Marchant, D. R., and G. H. Denton. 1996. Miocene and Pliocene paleoclimate of the Dry Valley region, southern Victoria Land: A geomorphological approach. Marine Mi-cropaleontology 27:253-271.

Mayewski, P. A., and 13 others. 1996. Climate change during the last deglaciation in Antarctica. Science 272:1636-1638.

Mayewski, P.A., L. D. Meeker, M. S. Twickler, S. I. Whitlow, Q. Yang, W. B. Lyons, and M. Prentice. 1998. Major features and forcing of high-latitude northern hemisphere atmospheric circulation using a 110,000 year long glaciochemical series. Journal of Geophysical Research 102:26345-26366.

McKay, C. P., G. Clow, R. A. Wharton, Jr., and S. W. Squyres. 1985. Thickness of ice on perennially frozen lakes. Nature 313:561-562.

McKay, C. P., M. T. Mellon, and E. I. Friedmann. 1998. Soil temperatures and stability of ice-cemented ground in the McMurdo Dry Valleys, Antarctica. Antarctic Science, 10:31-38.

McKnight, D. M., D. K. Niyogi, A. S. Alger, A. Bomblies, P. A. Conovitz, and C. M. Tate. 1999. Dry valley streams in Antarctica: Ecosystems waiting for water. BioScience 49:985-995.

Miller, L. G., and G. R. Aiken. 1996. Effects of glacial meltwater inflows and moat freezing on mixing in an ice-covered Antarctic lake as interpreted from stable isotope and tritium distributions. Limnology and Oceanography 41:966-976.

Moorehead, D. L., P. T. Doran, A. G. Fountain, W. B. Lyons, D. M. McKnight, J. C. Priscu, R. A. Virginia, and D. H. Wall. 1999. Ecological legacies: Impacts on ecosystems of the McMurdo Dry Valleys. BioScience 49:1009-1019.

Morgan, V. I., C. W. Wolley, J. Li, T. D. van Ommen, W. Skinner, and M. F. Fitzpatrick. 1997. Site information and initial results from deep drilling on Law Dome, Antarctica. Journal of Glaciology 48:3-10.

O'Brien, S. R., P. A. Mayewski, L. D. Meeker, D. A. Meese, M. S. Twickler, and S. I. Whitlow. 1995. Complexity of Holocene climate as reconstructed from a Greenland ice core. Science 270:1962-1964.

Parker, B. C., G. M. Simmons, Jr., G. F. Love, R. A. Wharton, Jr., and K. G. Seaburg. 1982. Removal of organic and inorganic matter from Antarctic lakes by aerial escape of blue-green algal mats. Journal of Phycology 18:72-78.

Petit, J. R., and 18 others. 1999. Climate and atmospheric history of the past 420,000 years from the Vostok ice record, Antarctica. Science 286:2141-2144.

Priscu, J. C. 1995. Phytoplankton nutrient deficiency in lakes of the McMurdo dry valleys, Antarctica. Freshwater Biology 34:215-227.

Priscu, J. C. (Editor). 1998. Ecosystem dynamics in a polar desert, the McMurdo Dry Valleys, Antarctica. Antarctic Research Series 72. American Geophysical Union, Washington, D.C.

Priscu, J. C., C. H. Fritsen, E. E. Adams, S. I. Giovannoni, H. W. Paerl, C. P. McKay, P. T. Doran, D. A. Gordon, B. I. Lanoil, and J. L. Pinckney. 1998. Perennial Antarctic lake ice: An oasis for life in a polar desert. Science 280:2095-2098.

Priscu, J. C., C. F. Wolf, C. D. Takacs, C. H. Fritsen, J. Laybourn-Parry, E. C. Roberts, B. Sattler, and W. B. Lyons. 1999. Carbon transformations in a perennially ice-covered Antarctic lake. BioScience 49:997-1008.

Roberts, E. C., and J. Laybourn-Parry. 1999. Mixotrophic cryptophytes and their predators in the Dry Valley lakes of Antarctica. Freshwater Biology 41:737-746.

Spigel, R. H., and J. C. Priscu. 1998. Physical limnology of the McMurdo Dry Valleys lakes. Pages 153-187 in J. C. Priscu, editor, Ecosystem dynamics in a polar desert: The McMurdo Dry Valleys, Antarctica. Antarctic Research Series 72, American Geophysical Union, Washington, D.C.

Steig, E. J., E. J. Brook, J. W. C. White, C. M. Sucher, M. L. Bender, S. J. Lehman, E. D. Waddington, D. L. Morse, and G. D. Clow, 1998. Synchronous climate changes in Antarctica and the North Atlantic. Science 282:92-95.

Steig, E. J., D. L. Morse, E. D. Waddington, M. Stuiver, P. M. Grootes, P. A. Mayewski, M. S. Twickler, and S. I. Whitlow. 2000. Wisconsinan and Holocene climate history from an ice core at Taylor Dome, Western Ross Embayment, Antarctica. Geografiska Annaler 82A: 213-235.

Stuiver M., G. H. Denton, T. J. Hughes, and J. L. Fastook. 1981. History of the marine ice sheet in West Antarctica during the last glaciation: A working hypothesis. Pages 319-436 in G. H. Denton and T. H. Hughes, editors. The last great ice sheets. Wiley-Interscience, New York.

Swanson, F. J., T. K. Kratz, N. Caine, R. G. Woodmansee. 1988. Landform effects on ecosystem patterns and processes. BioScience 38:92-98.

Tyler, S. W., P. G. Cook, A. Z. Butt, J. M. Thomas, P. T. Doran, and W. B. Lyons. 1998. Evidence of deep circulation in two perennially ice-covered Antarctic lakes. Limnology and Oceanography 43:625-635.

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

Virginia, R. A., and D. H. Wall. 1999. How soils structure communities in the Antarctic dry valleys. BioScience 49:973-983.

Vogt, E. A. 1997. Ecosystems: Balancing science with management. Springer-Verlag, New York.

Wharton, R. A., Jr., C. P. McKay, G. D. Clow, and D. T. Andersen. 1993. Perennial ice covers and their influence on Antarctic lake ecosystems. Pages 53-70 in W. J. Green, and E. I. Friedmann, editors. Physical and biological processes in Antarctic lakes. Antarctic Research Series 59, American Geophysical Union, Washington, D.C.

Wharton, R. A., Jr., C. P. McKay, B. C. Parker, and G. M. Simmons, Jr. 1986. Oxygen budget of a perennially ice-covered Antarctic dry valley lake. Limnology and Oceanography 31:437-443.

Wharton, R. A., Jr., C. P. McKay, G. M. Simmons, Jr., and B. C. Parker. 1985. Cryoconite holes on glaciers. BioScience 35:499-503.

Wilch, T. R., G. H. Denton, D. R. Lux, and W. C. Mcintosh. 1993. Limited Pliocent glacier extent and surface uplift in Middle Taylor Valley, Antarctica. Geografiska Annaler 75A:331-351.

Wilson, A. T. 1964. Evidence from chemical diffusions of climatic changes in the McMurdo Dry Valley 1200 years ago. Nature 201:176-177.

0 0

Post a comment