Research Hypotheses

MCM6 is focused on the roles of legacy and ecological connectivity in a polar desert ecosystem.

Overarching Hypothesis

The structure and functioning of the McMurdo Dry Valley (MDV) ecosystem is dependent upon legacies and the contemporary frequency, duration, and magnitude of ecological connectivity.

In MCM6, we consider the ecosystem consequences of temporal variability in connectivity, including connections occurring with low to high frequency, weak to strong magnitude, and short to long duration. Aeolian connectivity is often short duration (foehn storms occur for <1 day to a few days), infrequent across wide scales (~monthly), and variable in magnitude but more consistent in winter. Streamflow, in contrast, is of longer duration (weeks to months, only in summer) and varies in magnitude depending on meltwater generation. During warmer summers, high connectivity can occur for several weeks in the summer between the lakes, streams, and soils through the melting of perimeter moats. Abiotic and biotic attributes of the MDV ecosystem also have predictable annual rhythm. Six months of polar night occurs every year and the light regime varies little (except for cloudiness) in the summers, and air temperatures plunge far below freezing in the austral winters and rise a few degrees above freezing in the austral summers (though soil surfaces can warm to 20°C). These daily, seasonal, and annual events occur within a context of millennial-scale climate variation that includes both cooling and warming, which elicit significant responses of the mass balance of glaciers and lakes. Punctuating these time spans and degrees of connectivity are extreme weather events, such as the flood of 2001-02, which can significantly alter the existing structure and functioning of the MDV ecosystem. However, the resident biota in these ecosystems have evolved under these extreme regimes of energy and light input. The landscape has also developed from geologic, glacial, and prior ecological processes that have left legacies of minerals, organic matter, nutrients, and widely varying exposure ages.

To address this overarching hypothesis and to test the dependency of these systems on the frequency, duration, and magnitude of ecological connectivity, we propose the following four working hypotheses that integrate our proposed experiments and monitoring activities.

Working Hypotheses

H1: The stability of the MDV ecosystem with respect to contemporary variation in abiotic drivers is determined by the responses of sentinel taxa.

We define sentinel taxa as those which often exhibit high sensitivity to environmental disturbance, such that surveillance of these taxa can be used to assess ecosystem health. For example, in the MDVs, nematode species respond to changes in water availability in predictable ways and are apex level consumers responsible for ~10% of carbon turnover. Similarly, several phytoplankton taxa are responsible for the vast majority of autochthonous carbon in MDV lakes and streams. Thus, responses of sentinel taxa to changes in abiotic drivers can alter major ecological processes. Ecological resilience is a challenging concept to define, and to test. Our answer is to measure multiple aspects of diversity (richness, composition, functional) and ecosystem processes (gene expression, extracellular enzyme activity, respiration, and photosynthetic pigment activity) over multiple time scales. This will allow us to assess the resilience of MDV sentinel taxa to the abiotic variation associated with a changing climate, and to identify linkages between these organismal and population-level responses to resilience at an integrated community and ecosystem level. To test H1, we have developed surveillance activities of sentinel taxa in each landscape unit. We predict that the resilience of aquatic and terrestrial ecosystems to environmental disturbance will correlate with the changes in diversity that result from sentinel taxon responses to disturbance.

H2: The resilience of the MDV ecosystem to disturbance is dependent upon ecological legacy.

In MCM2, we developed a legacy model that linked geological history and paleoclimate to contemporary ecosystem structure and functioning. The current spatial distribution of nutrients, organic matter, biomass and organisms is influenced by ancient physical drivers. Some of these legacy drivers include soil salt accumulations and valley-filling paleolake deposits of ancient organic matter in soils, creating gradients of nitrogen and phosphorus availability among the different MDV basins that constrains biological activity. Other legacies include the chemoclines and biomass accumulation of the highly stratified lakes and glacier mass-balance, which influences stream flow. These ecological legacies, which arose and have persisted since the Neogene, overprint the different MDV landscape features and are powerful drivers of contemporary ecosystem structure and functioning. Testing hypotheses about how MDV ecosystems respond to contemporary climate variability (H1, H3, H4) requires that our observations be interpreted in the context of legacies persisting, despite observed contemporary changes in physical and biological connectivity. For example, even massive changes in contemporary hydrological connectivity may be insufficient to alter legacy drivers of diversity and distribution, and subsequent ecosystem structure and functioning. Thus, H2 predicts that the persistent effects of ecological legacy play a key role in maintaining ecosystem resilience to contemporary climate-driven changes. Importantly, H2 distinguishes ecological legacy from the contemporary drivers of ecosystem responses of the experiments and observations that are explicitly tested by H1, H3, and H4.

H3: Carryover of water, organic matter, and nutrients through periods of low connectivity maintains biological activity and community stability.

In desert and drought-prone ecosystems, biological activity can rebound quickly following water pulses after a dry period. In the MDVs, we know that the availability of resources for resident biota vary over multiple time scales, with contrasting dynamics in soil, stream, and lake landscape units. Most obviously, hydrologic connectivity to streams and soils is cut off through the winter months when glacial meltwater is no longer generated, promoting the desiccation of microbial mats, and soil biota suspend biological activities until favorable conditions return the next summer. However, shorter durations of hydrologic disconnection also occur in summer months during periods of lower insolation when soil water freezes. In soils, over 50% of the metazoan community exist in anhydrobiotic states even during the summer months, but remain capable of rapid revival upon pulses of rehydration. During daily and weekly periods of low water availability and reduced resource influx, biological activity must rely on local nutrients, organic matter, and water. In contrast, lake biota are not water-limited, but do rely on PAR and an influx of limiting nutrients from streams and wind-borne dust. Intermittent aeolian connectivity across landscape units persists through periods of hydrologic disconnectivity, supporting local resource influx. Overall, we predict that the amount of surplus resources accumulated at a locality are proportional to the level of biological activity maintained through periods of disconnection, and to the level of biological activity that is achieved upon reconnection. We will evaluate this hypothesis in MDV streams, soils, and lakes both by monitoring biological activity at a higher temporal resolution during focused times spanning disconnection and reconnection periods, and by experimentally altering the magnitude, frequency, and duration of resource supply or biologically favorable conditions.

H4: Changes in disturbance magnitude, frequency, and/or duration disrupt the annual reset of major ecological cycles following the polar night.

Winter conditions are rapidly changing in polar and temperate ecosystems, particularly those with significant seasonal snow and ice cover. As a result, changing light, temperature extremes, snowfall, wind, and timing of these conditions can have significant and lasting influences over biotic communities and ecosystem processes during the subsequent season. The polar night in the MDVs is a period of 6 months of darkness, which is experienced differently by communities across the landscape. In soil and stream habitats, communities and are presumably dormant at the persistent < -40°C temperatures. Lake moats freeze from the top down, resulting in a gradient from shallow moat constituents that freeze prior to loss of PAR to deep moat constituents that only freeze after months immersed in darkness. Liquid water persists year-round in the limnetic zones, while light-driven primary production ceases during the winter, representing a period of poorly understood heterotrophic and chemolithotrophic processes. We propose an extended field season to understand how environmental disturbance influences annual reset or resumption of ecological processes (e.g., primary productivity, decomposition, predation) following the polar night. Specifically, events that occur during the fall shut-down impact both winter activity and the state of the ecosystem at the beginning of the spring start-up. To fully address H4, which focuses on annual ecological cycles and carryover through the polar night, we must both develop robust baselines for annual ecological and hydrological cycles beyond the traditional Antarctic field deployment season and test the influence of disturbances on ecosystem functioning and recovery into the fall/winter. To this end, we propose to conduct extended field work during the 2026-27 season into April of 2027 (i.e., capturing the polar night transition), as the traditional Antarctic field season is completed before the shutdown of hydrological activity in the MDVs, which is known to occur in March. It is unknown, however, how these ecosystems typically “shut down” during this transition into polar night, which requires detailed monitoring to understand annual ecological cycles and their relation to environmental variables. Experiments conducted during the extended season will directly address H4 by examining how processes during seasonal shutdown propagate through the polar night, into the start of the following summer. We will extend monitoring and experiments into a late-summer and early-autumn shoulder season and we will further test H4 by conducting manipulative field and laboratory experiments, remote sensing, and year-round automated sampling. We predict that environmental disturbances from the previous season that significantly impact the communities will disrupt natural processes which occur during the polar night