Abstract
Climate-driven glacial melt is altering polar ecosystems. Shifts in hydrological regimes have cascading effects on limno-terrestrial ecosystems. In the McMurdo Dry Valleys (Southern Victoria Land, Antarctica), year-round ice cover isolates lentic habitats, yet seasonal melt along the lake perimeter forms open-water “moats” during the short austral summer provide transient hydrological connectivity among soils, benthos, and the stratified water columns of the dry valley lakes. To investigate how connectivity influences biological communities, we tracked biodiversity, phytoplankton photosynthesis, and physicochemistry along lateral transects in two McMurdo Dry Valley lakes, Fryxell and Bonney. These lakes, shaped by distinct basin features (bathymetry, streams) and ecological legacies (nutrient status, chemistry), exhibited contrasting degrees of limno-terrestrial connectivity. Our data reveal that lake-specific hydrological linkages restructure microbial and invertebrate communities. We conclude that climate-induced hydrological changes destabilize previously stratified systems, altering ecological interactions and fundamental ecosystem processes across Antarctic limno-terrestrial ecosystems. Our findings provide critical insight into how polar freshwater ecosystems may reorganize under future climate scenarios, informing predictions of microbial community resilience in extreme environments.
Introduction
In recent decades, climate-driven environmental change has been proceeding at accelerated rates, increasingly affecting ecosystems at local and global scales (Pachauri and Meyer, 2014; Steffen et al., 2015; Turner et al., 2020). Climate change is also impacting global hydrological cycles, ranging from increases in annual precipitation, intense flooding events, modification in lake ice cover, and nutrient loading in lakes (Haddeland et al., 2014; Woolway et al., 2020; Hu et al., 2020). While many of the changes to ecosystems can be attributed to anthropogenic activity, there are ongoing challenges to delineate between predictable responses to environmental drivers and those associated with climate-driven change (Schultz et al., 2022).
Lakes represent diverse, geographically distributed ecosystems that play important roles in global hydrological and nutrient cycles (Adrian et al., 2009). Climate-driven environmental change is re-shaping freshwater systems globally—lake levels are rising (e.g., +0.2–0.4 m) in the North American Great Lakes by 2040s (Kayastha et al., 2022), while ice cover has decreased by ~25% in all five Great Lakes since 1970, shortening ice-covered seasons that alter phytoplankton dynamics and winter–summer cycles (Hampton et al., 2024). These shifts are driving nutrient loading, eutrophication, and tipping points across temperate and tropical lakes (Hessen et al., 2023). Understanding how climate variability impacts lakes is important in predicting future changes in lake functions.
The polar regions are especially sensitive to climatic-driven change. Rapid warming has been observed in the Arctic, Antarctic and the Tibetan Plateau during the second half of the twentieth century (Cauvy-Fraunié and Dangles, 2019; Šmejkalová et al., 2016). Dramatic changes in lake ice phenology in Arctic lakes (e.g., break-up/freeze-up dates; ice cover thinning, loss of perennial ice cover) are considered strong limnological indicators of climate change in the Northern Hemisphere (Brown and Duguay, 2011; Veillette et al., 2010; Wrona et al., 2016). Unlike in the Arctic, where surface air temperatures have increased more than twice the global average (Notz and Stroeve, 2016), warming in Antarctica is geographically patchy, with the western continent being most impacted, including large-scale mass loss of the Western Antarctic Ice Sheet (Bromwich et al., 2013; Joughin and Alley, 2011; Nicolas and Bromwich, 2011). Though recent analyses are showing warming over much of the continent (Nielsen et al., 2024; Tuckett et al., 2025; Zheng et al., 2025).
The McMurdo Dry Valleys (MDV) are located in Southern Victoria Land, East Antarctica, and represent the largest ice-free region on the Antarctic continent. Although the MDV are a polar desert, they contain numerous closed-basin lakes, representing biological oases that provide a rare source of year-round liquid water (Priscu, 1999). Each lake basin has a unique geomorphological and geochemical history which has legacy influences on current day physicochemical signatures of the water columns, including nutrient status, stream input, gases (Priscu et al., 2008), and lake bathymetry (Green and Lyons, 2009). As the lowest points in the watershed, the lakes are critical interfaces between the aquatic-terrestrial habitats of the MDV, integrating physicochemical and biological inputs from the surrounding basins.
Lakes serve as integrators of a region's hydrological cycle, and particularly in closed basins, lake-level records offer valuable insights into past climate conditions. Perennially ice-covered lakes are very common features of the polar deserts along the Antarctic coast: they can exhibit large variations in physicochemistry properties; for example, some are highly saline (e.g., Lake Bonney) and others draining directly into the ocean (e.g., Lake Miers). In the MDVs, lake levels across all documented systems have risen substantially at different rates, with periods of rapid rise punctuated by periods of relative stability (Gooseff et al., 2017; Morgan-Kiss et al., 2024). One of the most notable climate events in the region was an unprecedented flood between December 2001 and January 2002, triggered by a discrete warming episode that led to high rates of glacial melt and record-high stream discharge (Foreman et al., 2004). While the biological consequences of this flood were not fully characterized, existing data show that the event facilitated interactions between previously isolated communities, increasing connectivity across transition zones and contributing both biota and organic carbon to formerly dry, carbon-limited soils (Foreman et al., 2004; Fountain et al., 2016; Gooseff et al., 2017).
MDV lakes are ice-covered year-round, except for along the shoreline where open water, called “moats,” form for brief periods during the summer (i.e., December–January; Stone et al., 2024a,b). The soils adjacent to lakes are wetted throughout the summer and moats connect the lakes with the surrounding ecosystems and mediate hydrologic-limnologic interactions among glacial streams, arid soils, and the lakes (Gooseff et al., 2007), including wind-driven distribution of stream water (Castendyk et al., 2015). Moats are small yet highly dynamic habitats over spatial and temporal scales, distinct from the main lake water body, and disproportionately impacted by lake level rise (Stone et al., 2024a,b). To date, most studies in the MDV lakes have been intensely focused on interpreting data collected from the center of the lake. However, changes may be more dramatic in littoral zones where changes in ice duration and extent are more dramatic. Last, while benthic communities have been highly studied (Hawes, 2016; Hawes et al., 2014; Zhang et al., 2015), understanding of the diversity and function of the planktonic communities in the moats is currently limited.
In 2016, we conducted a Soil–Lake Inundation Moat Experiment (SLIME) to investigate ecological linkages across soils, moats, and lakes in response to climate-driven lake-level rise. In our first report from this initiative, we characterized the physicochemical conditions and seasonal dynamics of the open-water moat of Lake Fryxell, a shallow lake situated near the eastern end of Taylor Valley (Stone et al., 2024a,b). That work established a baseline understanding of moat development and identified the moat as a distinct and dynamic habitat influenced by seasonal meltwater input. Building on those findings, the current study expands the SLIME framework to compare two well-characterized MDV lakes—Fryxell and Bonney—which differ in their limnological histories, hydrological regimes, and bathymetric profiles. Here, we describe the bacteria, eukaryotes, and metazoans inhabiting the interconnected soil-moat–lake habitats at both sites. This broader approach provides a comparative framework to explore how lateral transitions between the moat and under-ice lake environments shape biodiversity and ecosystem structure in polar lake margins. Our study offers new insight into how biodiversity and biogeochemical function emerge from the interplay of spatial connectivity and ecological memory, providing clues to the resilience of polar desert ecosystems under a changing climate.
Methods
Site description and sample collection
Studies were conducted on Lakes Fryxell and Bonney which lie within the Taylor Valley (Figures 1A–C). Lake Fryxell, at the eastern part of the valley, has a surface area of 7 km2, and a maximum depth of 18 m. Lake Bonney is situated at the head of the valley, with a surface area of approximately 4.4 km2 and a maximum depth of 40 m. Lake Bonney has two basins, connected by a narrow (~20 m wide), shallow (12 m) channel. Detailed descriptions of these lakes may be found in Green and Lyons (2009) and Priscu (1998). Lake Fryxell and Bonney are fed by nine and two glacial streams, respectively (Supplementary Figure S1).
Figure 1
(A–C) Map showing locations of the study sites located in the east lobe of Lake Bonney (ELB) and Lake Fryxell (FRX) Taylor Valley, McMurdo Dry Valleys, Antarctica. Satellite imagery obtained from Sentinel-2 on Nov 18th, 2019, courtesy of Copernicus/ESA. (D–G). Schematics of sampling sites. Sampling transects were established at Lake Fryxell (NFRX, SFRX) and the east lobe of Lake Bonney (NELB, SELB) during the 2017–18 and 2018–19 austral summers, respectively. Shading blue, open water moat; green, under ice moat; orange, under ice water column.
SLIME sites were established at the north and south sides of Lake Fryxell (NFRX, SFRX) and the east lobe of Lake Bonney (NELB, SELB) during the field seasons of 2017–18 and 2018–19, respectively (Supplementary Table S1; Figures 1D–G). A detailed description of the Lake Fryxell SLIME transects was also recently reported by Stone et al. (2024a). Summer sampling occurred along transects that began with soils adjacent to the lakes, into the open water moat and continued to the water column under permanent ice cover (Figures 1D–G). Soils were collected along transects established perpendicular to lake margins that included soils from near the water's edge to outside the wetted zone to capture the chemical and physical variability described in Zeglin et al. (2009). Briefly, surface pebbles were removed when present and ~500 g soils were collected with sterile scoops into Whirlpak bags. Water samples from under the ice were collected through a 100 mm diameter hole drilled with a “Jiffy” ice drill. Water was collected using a peristaltic pump into acid-washed 1-L amber bottles. Benthic samples (microbial mats and underlying sediments) were collected along the transects by divers operating through holes melted in the ice cover. Benthic samples were collected using either a cut-off 60 mL syringe corer or spatulas into Whirlpaks or conical tubes. All samples were stored in coolers to protect them from freezing and light during transport to field laboratories located on the lake shores.
Soils were subsampled (~10 g) and preserved in sterile conical tubes by adding an equal mass of sucrose lysis buffer (SLB: 20 mM EDTA, 200 mM NaCl, 0.75 M sucrose, 50 mM Tris-HCl, pH 9.0; Giovannoni et al., 1990). Water samples (300 mL) were filtered through Millipore Sterivex-GP filters (Bedford, MA; pore size 0.22 μm) and preserved by adding 200 μL of SLB into the filter to cover the filter membrane. Benthic samples were frozen immediately upon returning to the laboratory, and then lyophilized and homogenized in McMurdo Station. Sample mass was determined before and after lyophilization. Lyophilized benthic samples were subsampled (0.2–1 g) and preserved with an equal mass of SLB. All subsamples were immediately frozen and stored at −20 to −80 °C until analyzed. The addition of SLB has been shown to effectively preserve community composition at ambient to low temperature (Mitchell and Takacs-Vesbach, 2008).
Physicochemistry
A suite of measurements was collected from the benthos and water column transects. For water column chlorophyll a (Chl-a), lake water (125 mL) was filtered onto 25 mm GF/F filters and extracted overnight in 90% acetone prior to analysis, which was performed on a Turner Designs Model 10-AU-005 fluorometer (Lizotte and Priscu, 1992). Temperature and conductivity were measured with either an SBE 25plus Sealogger CTD profiler (Sea-Bird Electronics Inc., WA) or a handheld CTD (YSI30). Nutrient and ion concentrations were measured using a Lachat Autoanalzer according to the McMurdo LTER limnology manual (http://mcm.lternet.edu/sites/default/files/MCM_Limno_Methods_AC_20171003.pdf).
Phytoplankton community composition and photosystem II (PSII) photochemistry
Chlorophyll fluorescence measurements were also conducted using the PHYTO-PAM II phytoplankton analyzer (Heinz Walz, Effeltrich, Germany). Before field deployment, algal references were constructed in the laboratory using pure cultures of Lake Bonney algae, Chlamydomonas sp. ICE-MDV (chlorophytes), Isochrysis sp. MDV (mixed/brown group), and an Antarctic cryptophyte, Gemigera cryophila. These references were then used to deconvolute the algal classes from field measurements. The PSII photochemical parameter FV/FM–maximum PSII photochemical efficiency was also measured for each algal class.
DNA extraction
DNA was extracted from all samples using a variation of the CTAB method described in Mitchell and Takacs-Vesbach (2008). Briefly, two volumes of CTAB buffer (1% CTAB, 0.75 M NaCl, 50 mM Tris pH 8, 10 mM EDTA) and proteinase K (final concentration 100 μg mL−1) were added to samples and incubated for 1 h at 60 °C on a continuous rotator. Sodium dodecyl sulfate was added to a final concentration of 2%, and samples were incubated for another 30 min on the rotator. DNA was extracted one time with an equal volume of phenol:chloroform:isoamyl alcohol (50:49:1) followed by two extractions with an equal volume of chloroform. DNA was precipitated by the addition of 0.1 volume of 3 M sodium acetate and two volumes of 95% ethanol, followed by overnight incubation at −20 °C. The samples were then centrifuged for 45 min (~21,000 x g), washed in 70% ethanol, and resuspended in 10 mM filter-sterilized Tris buffer pH 8.0.
Molecular diversity analysis
Samples were prepared for dual-index paired-end amplicon sequencing of the small subunit (SSU) of rRNA genes (16S and 18S rRNA genes) using V6 universal bacterial primers 939F 5′ TTG ACG GGG GCC CGC ACA AG-3′ and 1492R 5′-GTT TAC CTT GTT ACG ACT T-3′ (16S rRNA gene) and 1391F 5′-GTA CAC ACC GCC CGTC-3′ and EukBR 5′-GTA CAC ACC GCC CGTC-3′ for eukaryotic 18S rRNA genes (Caporaso et al., 2012; Stoeck et al., 2010). Primers included overhang adapter sequences for compatibility with Illumina index and sequencing adapters. PCR was performed in triplicate using 5Prime Hot Master Mix and a 55 °C annealing temperature for 30 cycles. Amplicons were cleaned and normalized using the Sequelprep kit (Fisher Science, Cat No. A1051001) and indexed using the Nextera XT index kit following manufacturer's instructions. Indexed amplicons were combined and cleaned using the AMPure XP bead cleaning kit. Libraries were run on an Illumina MiSeq using the v3 reagent kit with 18% PhiX sequencing control DNA. Bacterial DNA sequence data were analyzed in QIIME 1.9 (Bolyen et al., 2019) after trimming, filtering (PHRED > 20; length > 250) and interleaving (using default settings) in Sickle (Joshi and Fass, 2011) and PANDAseq (Masella et al., 2012). Eukaryotic data were filtered, trimmed and interleaved using DADA2 in R (R Core Team, 2013). Taxonomic assignments were made using the pre-trained naive Bayesian classifier and the Silva database training set v.138. Indicator species analysis was performed using the indicspecies (De Cáceres et al., 2010) package in R using the multipatt command and r.g. function (point biserial correlation coefficient) for 9,999 permutations.
Metazoan community composition
Animals were extracted directly from fresh collected mats and sediments from 50 mL conical tubes using a modified sieving and sugar flotation-centrifugation method. Sample extracts were poured over a 400 μm sieve on top of a 38 μm sieve. The material on the top sieve was thoroughly washed with reverse osmosis (RO) tap water to ensure as many of the animals as possible would pass through the 400 μm sieve and be caught on the 38 μm sieve. All material collected on the 38 μm sieve was rinsed with RO water into a 50 mL conical centrifuge tube and then extracted using sugar density gradient centrifugation according to Freckman and Virginia (1998).
Data accessibility
Datasets used in this manuscript are available on the MCM LTER website (mcmlter.org) and the environmental Data Initiative (EDI) Data Repository (portal.edirerepository.org; Morgan-Kiss and Takacs-Vesbach, 2025). The DNA sequence data from this project are available from the Sequence Read Archive at NCBI under BioProject PRJNA1290386.
Results and discussion
Characteristics of soil-moat-lake habitats
To investigate how hydrological connectivity influences biological community structure and ecosystem functioning, we conducted lateral transect sampling across the contrasting moat–lake systems of Lakes Fryxell and Bonney (Figure 1). Although MDV lakes have been the focus of intensive study for over three decades, most water column sampling has relied on repeated measurements from a single central ice hole, typically at the lake's deepest point. However, recent work by Stone et al. (2024b) highlighted substantial lateral heterogeneity in the moat of Lake Fryxell, revealing dynamic physicochemical gradients and seasonal variation in this transitional habitat. Following Stone et al. (2024a),b we define the moat as the part of the lake that largely melts in summer, under the influence of solar radiation, and largely freezes in winter. Moats melt from the bottom upwards, and residual ice covers, a few cm thick persist over some moat habitat, though the extent varies spatially and temporally. Building on that foundation, we expanded our spatial scope to assess biological communities along north and south shore transects in both Lake Fryxell and the east lobe of Lake Bonney (Figures 1D–G). We characterized microbial (bacterial and eukaryotic) diversity using 16S and 18S rRNA amplicon sequencing and enumerated benthic and planktonic metazoans across soil, water column and benthic habitats.
At the time of sampling in January 2017, NFRX possessed an extensive open moat, susceptible to wind-driven mixing extending to a depth of approximately 1 m where an ice-wall reached to the lake floor. Glacial streams nearest to the NFRX site are Canada and Huey (Conovitz et al., 2006). The permanent ice-cover was grounded to the lakebed, physically separating the open water moat from the stratified water column under the ice cover (Figure 1D). In contrast with NFRX, the moat habitat on SFRX remained completely ice covered (but the ice was not grounded), with little to no connection to the atmosphere (Figure 1E). Glacial streams nearest to the SFRX site are Von Guerard, Crescent and Harmish (Supplementary Figure S1). We established sampling transects at both north and south sites which included the open moat and adjacent stratified water column at NFRX (Figure 1D, blue and orange sites, respectively) and spanned the ice-covered moat at SFRX (Figure 1E, green sites). Total stream discharge during the sampling years was generally around the mean (Supplementary Figure S1).
North and south sites were sampled in Lake Bonney, east lobe (NELB, SELB, respectively) during January 2018. In contrast with Lake Fryxell, both NELB and SELB exhibited substantial hydrological connectivity between the moats and the ice-covered, stratified water column (Figures 1F, G). Both sites also exhibited open water moats at the lake margin and ice-covered moats further offshore. Owing to the steep bathymetry of the Lake Bonney basin, the ice-free zone of the NELB and SELB moats was narrower compared with NFRX. ELB is fed by only two streams (Priscu and Bohner): total discharge was a magnitude lower in ELB vs. FRX (Supplementary Figure S1). We sampled the open water moats, ice covered moats, as well as the adjacent stratified water column at both NELB and SELB (Figures 1F, G; blue, green and red sites, respectively).
Conductivity within the ice-covered water column of NFRX ranged from 600 μS in the water immediately below the lake ice to >1,000 μS, indicating that these sites were chemically stratified. By comparison, the conductivity of the NFRX open moat was notably fresher (104 μS) compared with adjacent surface waters on the other side of the grounded zone, supporting the observation that there was minimal hydrological connectivity between the moat and ice-covered water column. Conductivities of SFRX were consistently very low in all sampling holes (~ 15 μS; Supplementary Table S1), indicating that the liquid water originated from melting ice and/or streamflow from dilute glacial meltwater streams.
In contrast to Lake Fryxell, the water columns from sampling holes located adjacent to the open moats of either NELB or SELB exhibited no chemical stratification: conductivity values from Hole 2 which was adjacent to the moats was similar to that of the moats (Supplementary Tables S1, S2). However, Hole 3, which was located further away from the moat, exhibited typical chemical stratification at both NELB and SELB, as evidenced by an increase in conductivity with sampling depth (Supplementary Tables S1, S2). These results reveal that hydrological mixing in Lake Bonney was limited to the water column directly adjacent to the moats. Thus, the moats and ice-covered water columns of Lakes Fryxell and Bonney exhibited varying degrees of interactions within the lakes' regions and their surrounding ecosystem and provide an opportunity to observe planktonic and benthic communities exposed to variable hydrologic connectivity. These contrasting patterns between Lakes Fryxell and Bonney likely reflect not only current hydrological connectivity, but also legacy influences such as nutrient histories, basin bathymetry, and stream inflow dynamics established over millennial timescales.
Lake margins represent sites of future benthic communities
As lake levels in the region rise in response to glacial melt and stream inflow, physical and chemical legacies will imprint on newly forming benthic communities. We hypothesize that these communities are seeded from a combination of soil biota and lake communities. The distinct bacterial communities associated with the intermittently wetted soils of lake, stream and snowpacks within the MDV (Reynebeau, 2021; Van Horn et al., 2013; Zeglin et al., 2013) margins host a distinct bacterial community that varies with the conductivity and moisture profiles that result around these wetted zones. Specifically, bacteria in wetted zones are similar to communities found in other less arid polar and alpine zones while the composition of the dry soils adjacent to these hydrological margins resemble bacteria from disturbed sites (Zeglin et al., 2011). The eukaryotic soil communities described from MDV soils comprise described endemic species adapted to the extreme aridity, low biomass and high conductivity of the region (Adams et al., 2006). Diversity and biomass are often significantly related to soil moisture and other edaphic characteristics; the strength and direction of these relationships are highly contextual and vary across geographic scale (Van Horn et al., 2013). In this study we aimed to systematically characterize the prokaryotic and eukaryotic communities found along Lake Fryxell and Bonney hydrological margins as a first step toward establishing a baseline community to be monitored as lake levels rise (Supplementary Figure S2). We found that the soil Bacteria of both lake margins are dominated by Proteobacteria which were twice as abundant at Lake Bonney than Lake Fryxell (42% compared to 21%). Firmicutes and Bacteroidetes were also abundant at Lake Bonney (25–33% and 8–15%, respectively) whereas Acidobacteria comprised a significant proportion of the Lake Fryxell bacterial community (~27%). Members of TM7, Acidobacteria, Cyanobacteria, Actinobacteria, Planctomycetes, Defferribacteres, Fusobacteria, TM6, Verucomicrobia, and Tenericutes comprised 3% or less of the community detected in Lake Bonney (Supplementary Figure S2A). In Lake Fryxell, which had a more complex bacterial community, there was a longer list of low abundance phyla that in addition to the groups listed for Lake Bonney, also included Firmicutes, Cyanobacteria, Thermi, Nitrospirae, Chlorobi, and others.
Eukaryotic community composition in soil sites between and within lakes was more variable compared to bacterial composition. At Lake Bonney's south and north sites, members of the Basidiomycota (6–41%), Ascomycota (4–15%), Phragmoplastophyta (6–13%), Ochromycota (7–70%), Rotifera (< 1–4%) and Cryptomonadales (1–4%) were most abundant. At Lake Fryxell, unassigned Eukaryotes (26–50%), Chlorophyta (12–21%), Ochrophyta (11–19%), Cercozoa (10–15%), Ciliophora (< 1–7%), Basidiomycota (2–3%), and Chytridiomycota (1–6%) were most abundant in the south and north sites, respectively (Supplementary Figure S2B). Based on these baseline data, biodiversity in lake margin soil transects is uneven, but includes phyla described previously for MDV soils (Van Horn et al., 2013; Zeglin et al., 2013). Future work will focus on monitoring compositional changes and relating them to lake level changes and existing benthic communities.
Connectivity drives phytoplankton diversity and function
To assess how hydrological connectivity influences phytoplankton diversity and function, we employed a spectral fluorometer, the Walz PHYTO-PAM-II, to capture phytoplankton diversity and photochemical activity. The proportion of major spectral classes of phytoplankton differed between the two lakes, regardless of sampling location. Lake Bonney communities were largely dominated by green algae (chlorophytes), while Lake Fryxell phytoplankton communities were more diverse (Figure 2).
Figure 2
Phytoplankton distribution in Lake Fryxell (A, B; NFRX, SFRX) and east lobe Lake Bonney (C, D; NELB, SELB). Abundance of spectral algal classes was determined on a Phyto PAM Instrument (Walz). Shading—blue, open water moat; green, under ice moat; orange, under ice water column.
Phytoplankton community structure and function in the moat and main lake water column reflected the degree on moat-main lake hydrological connectivity. In NFRX, phytoplankton communities sampled from the ice-covered water column were distinct from those residing in the open moat. The open moat of NFRX was composed of cyanobacteria, chlorophytes and a “mixed algae” community (note: previous studies identified haptophytes as the major contributors to this mixed community, Dolhi et al., 2015). Under the permanently ice-covered water column, cryptophytes became abundant and increased with depth, with maximum levels at the deepest sampling depth (Figure 2A). These spatial patterns agree with previous reports that Lake Fryxell has a robust cryptophyte population which reaches maximum levels in the permanent chemocline of the sampling hole in the middle of the lake (Kong et al., 2012; Li and Morgan-Kiss, 2019). In the ice-covered moat of SFRX, phytoplankton communities were different again and dominated by cyanobacteria (Figure 2B).
While the phytoplankton composition of Lake Fryxell sites within the main lake generally agreed with previous studies of phytoplankton composition in the center of the lake, cryptophytes were absent from moat samples in NFRX and SFRX. Further, in Lake Bonney phytoplankton composition differed from previous reports: lower proportions of cryptophytes were reported in all sampling sites of NELB and SELB in the current study (Figures 2C, D) than from samples taken from similar depths in the middle of the lake (Dolhi et al., 2015). These data suggest that the open water moats in both lakes and increased hydrological activity between moats and the main lake in Lake Bonney do not provide a suitable habitat for cryptophytes. In an incubation experiment which transplanted phytoplankton from the under-ice water column to the open water moats, Sherwell et al. (2022) observed that cryptophytes were highly sensitive to disturbance. Conversely, chlorophyte communities from the MDV lakes appear to be resilient to disturbance (Sherwell et al., 2022), including addition of nutrients (Teufel et al., 2017).
The spatial segregation observed between moat and adjacent under-ice communities in the main lake of the NFRX phytoplankton communities was absent from both NELB and SELB (Figures 2C, D). All sample sites in the moats and the stratified water columns were dominated by chlorophytes in NELB and SELB, while other algal groups were minimally present in Lake Bonney sampling sites. These data extend previous reports which have only sampled the stable habitat of the middle of Lake Bonney (Li and Morgan-Kiss, 2019; Dolhi et al., 2015): here we show that chlorophytes are a dominant phytoplankton taxon in communities residing in open moats and adjacent water column, where the environment is seasonally dynamic. A number of lab-based studies support this observation that MDV chlorophytes are adaptable to dynamic environments: Lake Bonney isolates Chlamydomonas priscui and Chlamydomonas sp. ICE-MDV exhibit dynamic abilities to tolerate and grow in the presence of a variety of environmental perturbations, including salinity stress and high light (Kalra et al., 2023; Popson et al., 2024).
We monitored maximum photosystem II photosynthetic efficiency (FV/FM) as an indicator of photosynthetic activity (Figure 3). In NFRX and SFRX, FV/FM levels were very low (< 0.1) to undetectable in the moats (Figures 3A, B). Within the stratified water column of NFRX, FV/FM values across the algal groups were higher compared to the moats, with most values >0.4 (Figure 3A). In Lake Bonney, FV/FM values were generally comparable for chlorophytes across all sampling sites, while other groups exhibited variability (Figures 3C, D). Previous studies on isolates from Lake Bonney have described several novel adaptations of the photosynthetic machinery in response to long-term exposure to stable light conditions (reviewed recently in: Morgan-Kiss et al., 2024). These results are the first documentation of photochemical activity in the native communities. Our findings indicate that the extreme conditions (i.e., nutrient deprivation, high light) in the moats largely inhibit photosynthetic activity. In contrast, multiple algal groups maintained photochemical activity within the under-ice communities. These results highlight the marked contrast in phytoplankton community activity between these adjacent habitats and the impact of ice-free conditions on the sensitive phytoplankton communities.
Figure 3
Phytoplankton photosynthetic efficiency (FV/FM) in Lake Fryxell (A, B; NFRX, SFRX) and The east lobe of Lake Bonney (C, D; NELB, SELB). Abundance of spectral algal classes was determined on a Phyto PAM Instrument (Walz). Shading—blue, open water moat; green, under ice moat; orange, under ice water column.
Hydrological connectivity and community structure of bacteria and eukaryotes
Planktonic community composition across transects
The results from 16S rRNA and 18S rRNA sequencing within the planktonic communities of the four transects showed some general similarities across the two lakes. Planktonic bacterial 16S rRNA gene communities in Lakes Fryxell and Bonney were dominated (average, range) by the Betaproteobacteria (49%, 0–74%), Bacteroidetes (29%, 7–57%), Cyanobacteria (14%, 4–40%), Planctomycetes (4%, 0–38%), Gammaproteobacteria (2%, 0–25%), and Verrucomicrobia (2%, 0–6%; Figure 4). Eukaryotic 18S rRNA genes were represented by members of the Cryptophyta (42%, 0.6–65%), Chlorophyta (20%, 1–62%), Ochrophyta (15%, 2–46%), as well as several others (Figure 5). These trends in overall bacterial and eukaryal community diversity agreed with previous studies focused on historical sampling holes in the middle of the lakes (Kwon et al., 2017; Li and Morgan-Kiss, 2019; Vick-Majors et al., 2014).
Figure 4
Distribution of Bacteria phyla from water column samples collected from north and south sampling sites of Lake Fryxell (A, B; NFRX, SFRX) and the east lobe of Lake Bonney (C, D; NELB, SELB). Shading—blue, open water moat; green, under ice moat; orange, under ice water column.
Figure 5
Distribution of Eukarya phyla from water columns samples collected from Lake Fryxell (A, B; NFRX, SFRX) and the east lobe of Lake Bonney (C, D; NELB, SELB). Shading—blue, open water moat; green, under ice moat; orange, under ice water column.
Relative distributions of the dominant bacterial and eukaryal populations varied between the two lakes, which we attribute to lake-specific variability in hydrological connectivity. In NFRX, where the moat was separated from the lake water column by a grounded ice barrier (Figure 1D), the bacterial and eukaryal communities in the moat were distinct from those of the water column under the permanent ice in the main lake (Figures 4A, 5A). Moat communities in NFRX had a greater abundance of Betaproteobacteria and Chlorophytes than the water column, while Cryptophytes, Bacteroidetes, Cyanobacteria, and Verrucomicrobia, and Planctomycetes were all more abundant in the planktonic communities of the permanently ice-covered water column of the main lake (Figure 4A). In SFRX, where all sampling sites from the moat were collected while the moat was covered by a lens of ice (Figure 1E), the bacterial and eukaryal communities were generally uniform across all samples and resembled the NFRX moat (Figures 4A, B, 5A, B). While Cryptophyta was a dominant eukaryote phylum in all NFRX samples, they were reduced to < 2% of the eukaryote communities in SFRX (Figure 5B). Actinobacteria were only detectable in samples from the under-ice planktonic communities (Figure 4).
Evidence of mixing across the moats and adjacent water column under the permanent ice cover was observed in Lake Bonney planktonic samples from both north and south sites (Figures 1F, G). Bacterial communities within the moats and the first sampling hole adjacent to the moats were dominated by Bacteroides and Betaproteobacteria (Figures 4C, D; blue, green and red sites). Actinobacteria and Planctomycetes were detectable only in the sample holes of NELB and SELB that were furthest away from the moats where the density gradient formed (Hole 3; Figures 4C, D; Supplementary Tables S1, S2). Similarly, eukaryote populations from the moats and the closest sampling hole exhibited comparable distributions of cryptophytes and chlorophytes (Figures 5C, D). Both sites exhibited a shift in the eukaryote communities in the stratified water column to favor higher proportions of cryptophytes and the presence of ciliates (Hole 3; Figures 5C, D).
The high abundance of 18S rRNA sequences identified as cryptophytes was distinct from the results of the FluoroProbe, most notably in Lake Bonney. This discrepancy between the two methods of characterizing phytoplankton diversity is interesting: it potentially reflects a functional phenomenon of this phytoplankton group. Specifically, cryptophytes are mixotrophic and may depend upon heterotrophy in the nutrient-poor waters of the dry valley lakes, functionally downregulating photosynthesis. There is also growing evidence in other freshwater environments that a significant proportion of cryptophytes are aplastidic, reaching levels as high as 75% of the heterotrophic nanoflagellate (HNF) populations (Šimek et al., 2023). Previous work has identified that HNF are a large proportion of the predator populations in Lake Bonney (Li and Morgan-Kiss, 2019). Further studies are needed to ascertain whether a significant portion of the dry valley lake cryptophytes are aplastidic and whether availability of light and other resources plays a role in loss of plastids.
We compared the top 25 ASVs across all four sampling sites. Two Betaproteobacteria related to Burkolderiales (ASV27 and ASV23) dominated all plankton samples from both Lakes Fryxell and Bonney. Several Bacteroides ASVs were also abundant and exhibited spatial trends across the lateral transects. A Flavobacteria (ASV10) and a Cytophagia (ASV7) were prevalent within the open water and ice-covered moats of NFRX and NELB and SFRX and SELB, respectively. A Sphingobacteria (ASV11) dominated the stratified water column of NFRX (Supplementary Figure S3). The planktonic eukaryote communities of all four sites were dominated by a single cryptophyte (ASV94; Supplementary Figure S3). Two chlorophyte ASVs (ASV71 and ASV73) exhibited spatial changes across the transects, with ASV71 being more prevalent in the moats (Supplementary Figure S3, blue and green sites). Two ochrophyte ASVs related to Chrysophyceae (ASV130 and ASV124) were detected in the ice-covered moat of SFRX, while an Ochromonas (ASV138) was evenly distributed across the moats and water columns of all four sites.
PCoA ordination analyses provided further support for the hydrological connectivity impacting microbial community structure. Bacteria in the open water and ice-covered moats of both lakes (Figure 6A, OM and IM samples) clustered tightly together, while most communities of the water column under the permanent ice formed a tight cluster. However, a few of the Lake Bonney water column samples clustered with the moat samples (Figure 6A, WC samples), indicating that the community structure of the Lake Bonney water column samples located adjacent to the moats resembled that of the moat communities. In contrast with the bacterial communities, the distinct eukaryote community within the ice-covered moat drove the majority of the clustering between NFRX and the rest of the sites (Figure 6B).
Figure 6
Principal coordinate analysis (PCoA) ordinations showing taxa scores (taxaplot) for Bacteria (A) and Eukarya (B) communities residing in Lakes Fryxell and Bonney.
Ordination of microbial community composition also revealed lake-specific structuring in both Bacterial and Eukaryotic assemblages (Figure 6). PCoA axis 2 (the Y-axis) showed a strong separation between samples from Lake Fryxell and Lake Bonney, indicating that overall community composition differs markedly between the two lake systems, consistent with previous findings (Li and Morgan-Kiss, 2019). Within each lake, additional clustering along PCoA axis 1 corresponded to depth-related variation, suggesting that vertical physicochemical gradients further shape community structure. These patterns were consistent for both domains, although bacterial communities showed greater dispersion along PCoA axis 1 compared to eukaryotic communities, indicating higher beta-diversity within lakes.
Correlation-based indicator species analysis of the planktonic communities revealed ASVs that were unique among the lakes and locations (Supplementary Tables S3, S4). Overall, Fryxell had at least twice as many unique organisms compared to Lake Bonney. When the communities of the moats, ice-covered moats and the water columns were compared, the ice-covered moats had the most distinct communities as a whole and by lake, for the 16S and 18S rRNA gene data. When the bacterial communities were considered by location in each of the lakes, the moats never had a community distinct from the ice-covered moat, suggesting this region of the lake was well-mixed. In addition, only Lake Bonney's moat region (moat + ice-covered moat) included any distinct community members, indicating a more homogenous composition among the transect from the moat to the water column. The 18S rRNA gene data of Lake Bonney was similarly homogeneous across the transects, though the water column also contained distinct members relative to the moat region.
Benthic microbial and metazoan responses to hydrological gradients and legacy effects
Benthic microbial communities in Lakes Fryxell and Bonney exhibited strong taxonomic differentiation from planktonic assemblages, with clear spatial structuring by lake and depth. Across all benthic samples, bacterial communities were dominated by Cyanobacteria (mean 31%, range < 1–72%), Betaproteobacteria (19%, 5–63%), Gammaproteobacteria (17%, 2–74%), Bacteroidetes (8%, 2–39%), Deltaproteobacteria (5%, < 1–51%), Acidobacteria (5%, < 1–25%), Planctomycetes (4%, < 1–11%), and Verrucomicrobia (3%, < 1–9%; Figures 7A, B; 8A, B). A similar separation of planktonic and benthic microbial assemblages was also reported from nearby Lake Vanda (Ramoneda et al., 2021). Notably, the abundance of dominant cyanobacterial taxa in Lake Fryxell benthic samples decreased with depth, while overall community heterogeneity increased (Figures 7A, B).
Figure 7
Distribution of major bacteria (A, B), Eukarya (C, D) phyla and animal counts (E, F) in benthic samples collected from Lake Fryxell (NFRX, SFRX).
Figure 8
Distribution of major bacteria (A, B), Eukarya (C, D) phyla and animal counts (E, F) in benthic samples collected from Lake Bonney.
Further taxonomic resolution showed that Fryxell benthic samples were enriched in Burkholderiales, Pseudomonadales, and Planctomycetota, particularly in nearshore cores (Supplementary Figure S4). In contrast, the deepest samples exhibited reduced evenness and diversity, likely reflecting the effects of oxygen limitation or sediment redox gradients. In Lake Bonney, benthic communities were comparatively enriched in Actinobacteriota and Alphaproteobacteria, with less dominance of Burkholderiales than in Fryxell (Supplementary Figure S4). Last, taxonomic overlap between benthic and water column samples was more pronounced in Bonney's nearshore sites, indicating greater hydrological connectivity and physical mixing.
Eukaryotic assemblages in the benthos were compositionally distinct from water column communities yet exhibited similar depth-dependent trends across both lakes. Shallow benthic samples from moat regions were dominated by rotifer sequences, while samples from deeper locations under the perennial ice cover consistently contained a higher relative abundance of diatoms (Figures 7C, D; 8C, D). Additional eukaryotic groups were present at lower abundance in all benthic samples, including chlorophytes, cryptophytes, ochrophytes, and dinoflagellates (Figures 7B, 8B). The presence of phototrophs and grazers in both shallow and deep sediments suggests vertical structuring of benthic food webs and supports prior observations of persistent microbial mat activity under low-light conditions (e.g., Hawes et al., 2014).
While the water column of the MDV lakes is largely devoid of higher life forms, benthic mats harbored diverse and abundant invertebrate communities. Invertebrate densities were consistently higher in mats than in underlying sediments, with notable taxonomic and spatial patterns across lakes and depths. This difference in nematode densities is driven by high microbial biomass content in the active mat relative to the underlying sediment (Stone et al., 2024a). The nematodes Scottnema lindsayae, Eudorylaimus sp., Plectus frigophilus, along with the tardigrade Acutuncus antarcticus, were more abundant in Lake Fryxell, consistent with its relatively stable and hydrologically isolated conditions (Figures 7E, F). In contrast, rotifers were more prevalent in Lake Bonney, particularly in shallower zones. This pattern suggests a response to greater hydrological connectivity and nutrient inputs. Across both lakes, the abundance of P. frigophilus and rotifers declined with depth, reflecting environmental filtering along gradients of light, moisture, and sediment geochemistry (Figures 7E, F, 8E, F).
Metazoan patterns offer ecological insight into both the contemporary function and historical stability of lake margin ecosystems. Nematodes such as Scottnema lindsayae and Plectus frigophilus, along with the tardigrade Acutuncus antarcticus, are closely linked to microbial mats through trophic interactions and habitat dependence. Their elevated abundances in Lake Fryxell, particularly in shallow, oxygenated mats, coincide with zones of high microbial diversity and phototrophic activity (Figures 7A–D, 8A, B), suggesting a coupling between metazoan distribution and microbial productivity. In contrast, the dominance of rotifers in Lake Bonney's shallower zones points to greater hydrological mixing and nutrient flux, which is also supported by the similarity in electrical conductivity. As slow-dispersing, heterotrophic consumers sensitive to redox, moisture, and carbon availability, these invertebrates likely respond to both real-time biogeochemical conditions and longer-term lake-level dynamics. Their sharp decline in deeper, more anoxic sediments, where anaerobic taxa rise and photosynthetic activity wanes, reinforces their potential as indicators of sediment chemistry, microbial mat function, and limno-terrestrial ecosystem resilience in the face of climate-driven restructuring. The depth-dependent shifts in rotifer and nematode abundance likely reflect not only immediate physicochemical gradients, but also legacy effects of lake stratification and ice-cover stability. That these patterns differ between Bonney and Fryxell suggests that metazoans, like microbial communities, serve as sentinels of long-term hydrological restructuring in these polar desert ecosystems. Given the decadal persistence of ice-cover and episodic melt events in MDV history, metazoan communities provide useful biological records of both past isolation and recent connectivity, functions that microbial taxa alone may not fully capture. Together, microbial and metazoan responses across lateral and vertical gradients reveal how benthic communities track both immediate environmental forcing and longer-term ecological memory embedded in stratification and redox boundaries.
Antarctic lake margins are climate-sensitive ecotones
Our findings demonstrate that climate-driven shifts in hydrological connectivity are not merely physical disturbances, but ecological forces capable of reorganizing microbial communities across Antarctic lake margins. These margins—encompassing soils, moats, benthic zones, and the stratified water column—function as climate-sensitive ecotones, where biotic and abiotic exchanges are amplified by changing meltwater dynamics. We observed that increased hydrological connectivity enhances material and biological flux across previously stratified habitats, disrupting established ecological boundaries.
Phytoplankton communities in Lake Fryxell exhibited pronounced spatial heterogeneity linked to the degree of moat–water column separation, with cryptophytes showing marked sensitivity to disturbance, while chlorophytes remained resilient to varying levels of connectivity. In contrast, the connectivity of the Lake Bonney basin revealed less spatial segregation, supporting the role of hydrological integration in shaping community structure. Similarly, bacterial and eukaryal communities shifted along gradients of connectivity in Lake Bonney, underscoring the influence of both contemporary hydrology and legacy geochemical conditions. Finally, the spatial heterogeneity of the benthic communities points to a strong influence of depth, or more proximally, photosynthetically available radiation on the keystone members of this region of the lakes. Models that incorporate surface energy balance and water column heat fluxes project that Lake Bonney may experience at least seasonal ice free conditions as soon as 2030 which would dramatically change the amount of light penetrating these lakes (Obryk et al., 2019). Our results indicate that the community composition and productivity of these lakes will be significantly impacted as ice cover on these lakes decreases. Specifically, we expect that biodiversity will decline with a potential loss of endemic species and lake productivity will be dramatically decreased under conditions where these lakes are no longer perennially covered by ice.
In conclusion, our findings show that Antarctic lake margins are critical ecological transition zones, where legacy features and ongoing hydrological change interact to reshape biodiversity and ecosystem processes. This work highlights the importance of sampling open moats together with under-ice lake communities to understand and predict the cascading effects of climate change on fragile limno-terrestrial interfaces.
Statements
Data availability statement
The original contributions presented in the study are publicly available. This data can be found here: NCBI BioProject, PRJNA1290386; Environmental Data Initiative (EDI) Data Repository, portal.edirerepository.org.
Author contributions
RM-K: Supervision, Data curation, Writing – review & editing, Methodology, Funding acquisition, Conceptualization, Investigation, Writing – original draft. BA: Funding acquisition, Writing – original draft, Formal analysis, Writing – review & editing, Investigation, Project administration, Methodology. SP: Formal analysis, Writing – review & editing. IK: Writing – review & editing, Investigation, Methodology. SS: Methodology, Writing – review & editing, Investigation. JB: Writing – review & editing, Methodology, Investigation. RK: Writing – review & editing, Data curation. SD: Writing – review & editing, Methodology, Investigation, Conceptualization. PD: Investigation, Conceptualization, Funding acquisition, Writing – review & editing, Methodology. MG: Investigation, Funding acquisition, Writing – review & editing, Methodology. IH: Data curation, Investigation, Writing – review & editing, Conceptualization. DM: Conceptualization, Writing – review & editing, Funding acquisition. JP: Conceptualization, Investigation, Writing – review & editing. CT-V: Funding acquisition, Conceptualization, Data curation, Writing – review & editing, Formal analysis, Investigation, Writing – original draft.
Funding
The author(s) declared that financial support was received for this work and/or its publication. This work was supported by grants from the National Science Foundation for Long Term Ecological Research (#OPP-1637708 and−2224760). IH was supported by the New Zealand Ministry of Business Innovation and Enterprise (ANTA1801).
Acknowledgments
This work builds on decades of ecological research in the Dry Valleys, and we acknowledge the foundational contributions of Robert Wharton, Diana Wall, Ross Virginia, Berry Lyons, Andrew Fountain, and others in establishing protocols and interpretive baselines. The authors thank Hilary Dugan for assistance with generation of Figure 1A and Supplementary Figure S1. The authors thank McMurdo LTER students and technicians for field and lab assistance. We also thank Antarctic Support Contractors and Petroleum Helicopters International for logistical support.
Conflict of interest
The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declared that generative AI was not used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
Author disclaimer
Any opinions, findings, conclusions, or recommendations expressed here are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.
Supplementary material
The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/ffwsc.2025.1682442/full#supplementary-material
References
1
Adams B. J. Bardgett R. D. Ayres E. Wall D. H. Aislabie J. Bamforth S. et al . (2006). Diversity and distribution of Victoria Land biota. Soil Biol. Biochem.38, 3003–3018. doi: 10.1016/j.soilbio.2006.04.030
2
Adrian R. O'Reilly C. M. Zagarese H. Baines S. B. Hessen D. O. Keller W. et al . (2009). Lakes as sentinels of climate change. Limnol. Oceanogr.54, 2283–2297. doi: 10.4319/lo.2009.54.6_part_2.2283
3
Bolyen E. Rideout J. R. Dillon M. R. Bokulich N. A. Abnet C. C. Al-Ghalith G. A. et al . (2019). Reproducible, interactive, scalable and extensible microbiome data science using QIIME 2. Nat. Biotechnol.37, 852–857. doi: 10.1038/s41587-019-0209-9
4
Bromwich D. H. Nicolas J. P. Wilson A. B. Monaghan A. J. Lazzara M. A. Keller L. M. et al . (2013). Central West Antarctica among the most rapidly warming regions on Earth. Nat. Geosci.6, 139–145. doi: 10.1038/ngeo1671
5
Brown L. Duguay C. (2011). The fate of lake ice in the North American Arctic. Cryosphere5, 869–892. doi: 10.5194/tc-5-869-2011
6
Caporaso J. G. Lauber C. L. Walters W. A. Berg-Lyons D. Huntley J. Fierer N. et al . (2012). Ultra-high-throughput microbial community analysis on the Illumina HiSeq and MiSeq platforms. ISME J.6, 1621–1624. doi: 10.1038/ismej.2012.8
7
Castendyk D. Mcknight D. Jaros C. Welch K. Niebuhr S. (2015). Pressure-driven, shoreline currents in a perennially ice-covered, pro-glacial lake in Antarctica, identified from a LiCl tracer injected into a pro-glacial stream. Hydrol. Processes29, 2212–2231. doi: 10.1002/hyp.10352
8
Cauvy-Fraunié S. Dangles O. (2019). A global synthesis of biodiversity responses to glacier retreat. Nat. Ecol. Evol.3, 1675–1685. doi: 10.1038/s41559-019-1042-8
9
Conovitz P. A. MacDonald L. H. McKnight D. M. (2006). Spatial and temporal active layer dynamics along three glacial meltwater streams in the McMurdo Dry Valleys, Antarctica. Arctic Antarctic Alpine Res.38, 42–53. doi: 10.1657/1523-0430(2006)038[0042:SATALD]2.0.CO;2
10
De Cáceres M. Legendre P. Moretti M. (2010). Improving indicator species analysis by combining groups of sites. Oikos119, 1674–1684. doi: 10.1111/j.1600-0706.2010.18334.x
11
Dolhi J. M. Teufel A. G. Morgan-Kiss R. M. Kong W. (2015). Diversity and spatial distribution of autotrophic communities within and between ice-covered Antarctic lakes (McM urdo Dry Valleys). Limnol. Oceanogr.60, 977–991. doi: 10.1002/lno.10071
12
Foreman C. M. Wolf C. F. Priscu J. C. (2004). Impact of episodic warming events. Aquatic Geochem.10, 239–268. doi: 10.1007/s10498-004-2261-3
13
Fountain A. G. Saba G. Adams B. Doran P. Fraser W. Gooseff M. N. et al . (2016). The impact of a large-scale climate event on Antarctic ecosystem processes. Bioscience66, 848–863. doi: 10.1093/biosci/biw110
14
Freckman D. W. Virginia R. A. (1998). “Soil biodiversity and community structure in the McMurdo Dry Valleys, Antarctica,” in Ecosystem Dynamics in a Polar Desert: The McMurdo Dry Valleys, Antarctica, Vol. 72 (Washington, DC: American Geophysical Society).
15
Giovannoni S. J. DeLong E. F. Schmidt T. M. Pace N. R. (1990). Tangential flow filtration and preliminary phylogenetic analysis of marine picoplankton. Appl. Environ. Microbiol.56, 2572–2575. doi: 10.1128/aem.56.8.2572-2575.1990
16
Gooseff M. N. Barrett J. E. Adams B. J. Doran P. T. Fountain A. G. Lyons W. B. et al . (2017). Decadal ecosystem response to an anomalous melt season in a polar desert in Antarctica. Nat. Ecol. Evol.1, 1334–1338. doi: 10.1038/s41559-017-0253-0
17
Gooseff M. N. Barrett J. E. Northcott M. L. Bate D. B. Hill K. R. Zeglin L. H. et al . (2007). Controls on the spatial dimensions of wetted hydrologic margins of two Antarctic lakes. Vadose Zone J.6, 841–848. doi: 10.2136/vzj2006.0161
18
Green W. J. Lyons W. B. (2009). The saline lakes of the McMurdo Dry Valleys, Antarctica. Aquatic Geochem.15, 321–348. doi: 10.1007/s10498-008-9052-1
19
Haddeland I. Heinke J. Biemans H. Eisner S. Flörke M. Hanasaki N. et al . (2014). Global water resources affected by human interventions and climate change. Proc. Natl. Acad. Sci. U.S.A.111, 3251–3256. doi: 10.1073/pnas.1222475110
20
Hampton S. Dugan H. Sadro S. Vick-Majors T. Ozersky T. (2024). Winter limnology on the rise. Limnol. Oceanogr. Bull.33:132. doi: 10.1002/lob.10643
21
Hawes I. (2016). Chlorophyll-A Concentrations in Lake Hoare Benthic Mats 1996-97. Environmental Data Initiative. doi: 10.6073/pasta/d3e9d47d479379d6b9c30b8ecaf48968
22
Hawes I. Giles H. Doran P. T. (2014). Estimating photosynthetic activity in microbial mats in an ice-covered Antarctic lake using automated oxygen microelectrode profiling and variable chlorophyll fluorescence. Limnol. Oceanogr.59, 674–688. doi: 10.4319/lo.2014.59.3.0674
23
Hessen D. O. Andersen T. Armstrong McKay D. Kosten S. Meerhoff M. Pickard A. et al . (2023). Lake ecosystem tipping points and climate feedbacks. Earth Syst. Dyn. Discuss.2023, 1–34. doi: 10.5194/esd-2023-22
24
Hu Y. Jiang X. Shao K. Tang X. Qin B. Gao G. (2020). Global lake responses to climate change. Nat. Rev. Earth Environ.1, 388–403. doi: 10.1038/s43017-020-0067-5
25
Joshi N. A. Fass J. N. (2011). Sickle: a Sliding-Window, Adaptive, Quality-Based Trimming Tool for FastQ Files (Version 1.33) [Software]. Available online at: https://github.com/najoshi/sickle (Accessed January 9, 2026).
26
Joughin I. Alley R. B. (2011). Stability of the West Antarctic ice sheet in a warming world. Nat. Geosci.4, 506–513. doi: 10.1038/ngeo1194
27
Kalra I. Wang X. Zhang R. Morgan-Kiss R. (2023). High salt-induced PSI-supercomplex is associated with high CEF and attenuation of state transitions. Photosynthesis Res.157, 65–84. doi: 10.1007/s11120-023-01032-y
28
Kayastha M. B. Ye X. Huang C. Xue P. (2022). Future rise of the Great Lakes water levels under climate change. J. Hydrol.612:128205. doi: 10.1016/j.jhydrol.2022.128205
29
Kong W. Ream D. C. Priscu J. C. Morgan-Kiss R. M. (2012). Diversity and expression of RubisCO genes in a perennially ice-covered Antarctic lake during the polar night transition. Appl. Environ. Microbiol.78, 4358–4366. doi: 10.1128/AEM.00029-12
30
Kwon M. Kim M. Takacs-Vesbach C. Lee J. Hong S. G. Kim S. J. et al . (2017). Niche specialization of bacteria in permanently ice-covered lakes of the McMurdo Dry Valleys, Antarctica. Environ. Microbiol.19, 2258–2271. doi: 10.1111/1462-2920.13721
31
Li W. Morgan-Kiss R. M. (2019). Influence of environmental drivers and potential interactions on the distribution of microbial communities from three permanently stratified Antarctic lakes. Front. Microbiol.10:1067. doi: 10.3389/fmicb.2019.01067
32
Lizotte M. Priscu J. (1992). Algal pigments as markers for stratified phytoplankton populations in Lake Bonney(dry valleys). Antarctic J. U.S.27, 259–260.
33
Masella A. P. Bartram A. K. Truszkowski J. M. Brown D. G. Neufeld J. D. (2012). PANDAseq: paired-end assembler for illumina sequences. BMC Bioinform.13, 1–7. doi: 10.1186/1471-2105-13-31
34
Mitchell K. R. Takacs-Vesbach C. D. (2008). A comparison of methods for total community DNA preservation and extraction from various thermal environments. J. Indus. Microbiol. Biotechnol.35, 1139–1147. doi: 10.1007/s10295-008-0393-y
35
Morgan-Kiss R. M. Popson D. Pereira R. Dolhi-Binder J. Teufel A. Li W. et al . (2024). Sentinel protist taxa of the McMurdo Dry Valley lakes, Antarctica: a review. Front. Ecol. Evol.12:1323472. doi: 10.3389/fevo.2024.1323472
36
Morgan-Kiss R. M. Takacs-Vesbach C. D. (2025). Soil Lake Inundation Moat Experiment (SLIME): Physical, Chemical, and Biological Measurements From Planktonic Water Columns, McMurdo Dry Valleys, Antarctica (2018–2020). Environmental Data Initiative. doi: 10.6073/pasta/2beced61dc25064c3197359d2f7967ff
37
Nicolas J. P. Bromwich D. H. (2011). Climate of West Antarctica and influence of marine air intrusions. J. Clim.24, 49–67. doi: 10.1175/2010JCLI3522.1
38
Nielsen E. B. Katurji M. Zawar-Reza P. Cullen N. J. (2024). Air temperature trends and extreme warming events across regions of Antarctica for the period 2003-2021. Authorea Preprints. 130:e2024JD043042. doi: 10.22541/essoar.171926315.55910608/v1
39
Notz D. Stroeve J. (2016). Observed Arctic sea-ice loss directly follows anthropogenic CO2 emission. Science354, 747–750. doi: 10.1126/science.aag2345
40
Obryk M. Doran P. Priscu J. (2019). Prediction of ice-free conditions for a perennially ice-covered Antarctic lake. J. Geophys. Res. Earth. ace124, 686–694. doi: 10.1029/2018JF004756
41
Pachauri R. K. Meyer L. A. (eds.). (2014). “IPCC: climate change 2014: synthesis report,” in Contribution of Working Groups I, II and III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change (Geneva: IPCC), 151.
42
Popson D. D'Silva S. Wheeless K. Morgan-Kiss R. (2024). Permanent stress adaptation and unexpected high light tolerance in the shade-adapted Chlamydomonas priscui. Plants13:2254. doi: 10.3390/plants13162254
43
Priscu J. C. (1998). Ecosystem Dynamics in a Polar Desert: The McMurdo Dry Valleys, Antarctica. Washington, DC: American Geophysical Union. doi: 10.1029/AR072
44
Priscu J. C. (1999). Life in the valley of the “dead”. Bioscience49, 959–959. doi: 10.2307/1313729
45
Priscu J. C. Dore J. E. Christner B. C. Westley M. B. Popp B. N. Casciotti K. L. et al . (2008). Supersaturated N2O in a perennially ice-covered Antarctic lake: molecular and stable isotopic evidence for a biogeochemical relict. Limnol. Oceanogr.53, 2439–2450. doi: 10.4319/lo.2008.53.6.2439
46
R Core Team (2013). R: A Language and Environment for Statistical Computing. Vienna: R Core Team. Available online at: http://www.R-project.org/
47
Ramoneda J. Hawes I. Pascual-García A. Mackey T. J. Sumner D. Y. Jungblut A. D. (2021). Importance of environmental factors over habitat connectivity in shaping bacterial communities in microbial mats and bacterioplankton in an Antarctic freshwater system. FEMS Microbiol. Ecol.97:fiab044. doi: 10.1093/femsec/fiab044
48
Reynebeau E. R. (2021). Quantification of soil microbial biomass in the McMurdo Dry Valleys, Antarctica (M.Sc. Thesis). University of New Mexico; UNM Digital Repository.
49
Schultz E. L. Hülsmann L. Pillet M. D. Hartig F. Breshears D. D. Record S. et al . (2022). Climate-driven, but dynamic and complex? A reconciliation of competing hypotheses for species' distributions. Ecol. Lett.25, 38–51. doi: 10.1111/ele.13902
50
Sherwell S. Kalra I. Li W. McKnight D. M. Priscu J. C. Morgan-Kiss R. M. (2022). Antarctic lake phytoplankton and bacteria from near-surface waters exhibit high sensitivity to climate-driven disturbance. Environ. Microbiol.24, 6017–6032. doi: 10.1111/1462-2920.16113
51
Šimek K. Mukherjee I. Szöke-Nagy T. Haber M. Salcher M. M. Ghai R. (2023). Cryptic and ubiquitous aplastidic cryptophytes are key freshwater flagellated bacterivores. ISME J.17, 84–94. doi: 10.1038/s41396-022-01326-4
52
Šmejkalová T. Edwards M. E. Dash J. (2016). Arctic lakes show strong decadal trend in earlier spring ice-out. Sci. Rep.6, 1–8. doi: 10.1038/srep38449
53
Steffen W. Deutsch L. Ludwig C. Broadgate W. Gaffney O. (2015). The trajectory of the Anthropocene: the great acceleration. Anthropocene Rev.2, 81–98. doi: 10.1177/2053019614564785
54
Stoeck T. Bass D. Nebel M. Christen R. Jones M. D. Breiner H. W. et al . (2010). Multiple marker parallel tag environmental DNA sequencing reveals a highly complex eukaryotic community in marine anoxic water. Mol. Ecol.19, 21–31. doi: 10.1111/j.1365-294X.2009.04480.x
55
Stone M. S. Devlin S. P. Hawes I. Welch K. A. Gooseff M. N. Takacs-Vesbach C. et al . (2024a). McMurdo Dry Valley lake edge ‘moats': the ecological intersection between terrestrial and aquatic polar desert habitats. Antarctic Sci.36, 1–17. doi: 10.1017/S0954102024000087
56
Stone M. S. Salvatore M. R. Dugan H. A. Myers M. E. Doran P. T. (2024b). Measuring and modelling functional moat area in perennially ice-covered Lake Fryxell, Antarctica. Arctic Antarctic Alpine Res.56:2406626. doi: 10.1080/15230430.2024.2406626
57
Teufel A. G. Li W. Morgan-Kiss R. M. Kiss A. J. (2017). Impact of nitrogen and phosphorus on phytoplankton production and bacterial community structure in two stratified Antarctic lakes: a bioassay approach. Polar Biol.40, 1007–1022. doi: 10.1007/s00300-016-2025-8
58
Tuckett P. A. Sole A. J. Livingstone S. J. Jones J. M. Lea J. M. Gilbert E. (2025). Continent-wide mapping shows increasing sensitivity of East Antarctica to meltwater ponding. Nat. Clim. Change15, 775–783. doi: 10.1038/s41558-025-02363-5
59
Turner M. G. Calder W. J. Cumming G. S. Hughes T. P. Jentsch A. LaDeau S. L. et al . (2020). Climate change, ecosystems and abrupt change: science priorities. Philos. Trans. R. Soc. B375:20190105. doi: 10.1098/rstb.2019.0105
60
Van Horn D. J. Van Horn M. L. Barrett J. E. Gooseff M. N. Altrichter A. E. Geyer K. M. et al . (2013). Factors controlling soil microbial biomass and bacterial diversity and community composition in a cold desert ecosystem: role of geographic scale. PLoS ONE8:e66103. doi: 10.1371/journal.pone.0066103
61
Veillette J. Martineau M. J. Antoniades D. Sarrazin D. Vincent W. F. (2010). Effects of loss of perennial lake ice on mixing and phytoplankton dynamics: insights from High Arctic Canada. Ann. Glaciol.51, 56–70. doi: 10.3189/172756411795931921
62
Vick-Majors T. J. Priscu J. C. Amaral-Zettler L. A. (2014). Modular community structure suggests metabolic plasticity during the transition to polar night in ice-covered Antarctic lakes. ISME J.8, 778–789. doi: 10.1038/ismej.2013.190
63
Woolway R. I. Kraemer B. M. Lenters J. D. Merchant C. J. O'Reilly C. M. Sharma S. et al . (2020). Global lake responses to climate change. Nat. Rev. Earth Environ.1, 388–403. doi: 10.1038/s43017-020-0067-5
64
Wrona F. J. Johansson M. Culp J. M. Jenkins A. Mård J. Myers-Smith I. H. et al . (2016). Transitions in Arctic ecosystems: ecological implications of a changing hydrological regime. J. Geophys. Res. Biogeosci.121, 650–674. doi: 10.1002/2015JG003133
65
Zeglin L. H. Bottomley P. J. Jumpponen A. Rice C. W. Arango M. Lindsley A. et al . (2013). Altered precipitation regime affects the function and composition of soil microbial communities on multiple time scales. Ecology94, 2334–2345. doi: 10.1890/12-2018.1
66
Zeglin L. H. Dahm C. N. Barrett J. E. Gooseff M. N. Fitpatrick S. K. Takacs-Vesbach C. D. (2011). Bacterial community structure along moisture gradients in the parafluvial sediments of two ephemeral desert streams. Microbial Ecol.61, 543–556. doi: 10.1007/s00248-010-9782-7
67
Zeglin L. H. Sinsabaugh R. L. Takacs-Vesbach C. D. Barrett J. E. Gooseff M. N. (2009). Landscape distribution of microbial activity in the McMurdo Dry Valleys: linked biotic processes, hydrology, and geochemistry in a cold desert ecosystem. Ecosystems12, 562–573. doi: 10.1007/s10021-009-9242-8
68
Zhang L. Jungblut A. D. Hawes I. Andersen D. T. Sumner D. Y. Mackey T. J. (2015). Cyanobacterial diversity in benthic mats of the McMurdo Dry Valley lakes, Antarctica. Polar Biol.38, 1097–1110. doi: 10.1007/s00300-015-1669-0
69
Zheng L. Liang Q. Cheng X. Shang X. van den Broeke M. R. Noël B. et al . (2025). Rapid increases in satellite-observed ice sheet surface meltwater production. Nat. Clim. Change15, 769–774. doi: 10.1038/s41558-025-02364-4
Summary
Keywords
Antarctica, ice-cover, lakes, microbial communities, primary production, benthic communities, lake margins
Citation
Morgan-Kiss RM, Adams BJ, Pothula SK, Kalra I, Sherwell S, Barrett JE, Korstvedt R, Devlin SP, Doran PT, Gooseff MN, Hawes I, McKnight DM, Priscu JC and Takacs-Vesbach C (2026) Climate-driven hydrological connectivity alters littoral and ice-covered ecosystems of Antarctic lake margins. Front. Freshw. Sci. 3:1682442. doi: 10.3389/ffwsc.2025.1682442
Received
08 August 2025
Revised
05 December 2025
Accepted
15 December 2025
Published
16 January 2026
Volume
3 - 2025
Edited by
Davide Taurozzi, Roma Tre University, Italy
Reviewed by
Qi Lu, Xiamen University, China
Manuel Castro Berman, Rensselaer Polytechnic Institute, United States
Updates
Copyright
© 2026 Morgan-Kiss, Adams, Pothula, Kalra, Sherwell, Barrett, Korstvedt, Devlin, Doran, Gooseff, Hawes, McKnight, Priscu and Takacs-Vesbach.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Rachael M. Morgan-Kiss, morganr2@miamioh.edu
† Present address: Isha Kalra, Department of Biology, University of Southern California, Los Angeles, CA, United States
Disclaimer
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.