Experimental cryoconite holes as mesocosms for studying community ecology

Abstract

Cryoconite holes are surface melt-holes in ice containing sediments and typically organisms. In Antarctica, they form an attractive system of isolated mesocosms in which to study microbial community dynamics in aquatic ecosystems. Although microbial assemblages within the cryoconite holes most closely resemble those from local streams, they develop their own distinctive composition. Here, we characterize the microbial taxa over time in cryoconite holes experimentally created from supraglacial sediments to demonstrate their utility as experimental mesocosms. We used high-throughput sequencing to characterize the assemblages of bacteria and microbial eukaryotes before melt-in, then after one and two months. Within one month of melt-in, the experimental holes, now lidded with ice, were visually indistinguishable from natural cryoconite holes, and within two months their thermal characteristics matched those of natural holes. The microbial composition of the experimental cryoconite holes declined in richness and changed significantly in the relative abundance of various taxa, consistent with possible turnover. In particular, a dominant cyanobacterium, Nostoc sp., further increased its dominance over the other dominant cyanobacterial phylotype, and an initially rarer Flavobacterium sp. became one of the dominant taxa. The eukaryotes continued to be dominated by algae and tardigrades, with the relative abundance of the dominant alga, Macrochloris sp., increasing notably relative to the microfauna. These changes within a single growing season in newly formed lidded cryoconite holes created from exposed supraglacial sediments are consistent with primary production and microbial turnover, and provide a promising foundation for future work using such mesocosms.

Introduction

Cryoconite holes are ice-bound water-filled holes that form on glaciers when sediments, termed cryoconite (Nordenskjöld 1875), melt into the glacial surface (Gribbon 1979; Wharton et al. 1985). They are common in snow-free areas on glaciers, covering 0.1–10% of the ablation zone of glaciers and ice sheets (Fountain et al. 2004; Anesio et al. 2009; Cook et al. 2012). Despite their small individual size, they are sufficiently numerous that the biogeochemical processes driven by microbial ecosystems within them can contribute significantly to overall carbon cycles (Anesio et al. 2009) and nutrient fluxes in oligotrophic regions such as the McMurdo Dry Valleys of Antarctica (Bagshaw et al. 2013).

Cryoconite holes develop when sediment is transported onto a glacier (Gribbon 1979) carrying with it bacteria, algae, fungi, and microfauna such as tardigrades and rotifers (Wharton et al. 1985; Cameron et al. 2012). The sediment, which has a lower albedo than the surrounding ice, absorbs solar radiation and melts into the glacier surface (Wharton et al. 1985; Fountain et al. 2004, 2008). Meltwater allows growth of microorganisms within the sediment, leading to communities that are often net photosynthetic (Bagshaw et al. 2016). The melting continues until the sediment layer is sufficiently deep that attenuation in radiation prevents it from melting any deeper, usually about 50 cm. At that point, melt approximately keeps pace with ablation of the ice surface (Gribbon 1979) and holes can become wider (Cook et al. 2010) in proportion to sediment volume (Cook et al. 2016).

In the McMurdo Dry Valleys, Antarctica, unlike Arctic and alpine regions, the summers remain sufficiently cold for cryoconite holes to retain a lid of ice over them, even as the sediment and a layer of water melt out within (Mueller et al. 2001). Although approximately half of the holes are connected by subsurface drainage channels (Fountain et al. 2004), many remain isolated and entombed in ice, resulting in extremely high pH values due to lack of gas exchange with the atmosphere in the presence of photosynthesis (Tranter et al. 2004; Webster-Brown et al. 2015). Distinctive and divergent microbial communities form within these isolated cryoconite holes (Stanish et al. 2013; Webster-Brown et al. 2015; Sommers et al. 2018). In Taylor Valley, one of the McMurdo Dry Valleys of Antarctica, microbial mats and sediments from the seasonal streams that flow from the glacier to the lakes have been suggested as major sources of cryoconite on the glaciers (Lancaster 2002; Stanish et al. 2013). Once the cryoconite hole forms, biological communities within them develop their own structure apart from the parent material in the streams. For instance, diatoms in cryoconite holes comprise only a subset of those found in the streams (Stanish et al. 2013), likely indicating different environmental conditions.

The purpose of our study was to determine the suitability of experimentally created cryoconite holes in Antarctica as mesocosms in which to test ecological theory. Previous work demonstrated that cryoconite holes can be artificially created by placing a thin patch of sediment on the glacial surface (MacDonell and Fitzsimons 2008). We hypothesized that the microbial communities present in sediments used to create cryoconite holes would undergo detectable changes in community structure within a single growing season as they grew in an environment distinct from that of streams or the surface of the glacier. To characterize changes, we sampled the communities in experimental holes after approximately one and two months after installation. Finally, to test whether experimental holes shared characteristics with natural holes, a replicate set of natural holes was selected for measurement of physical dimensions and community composition, and the temperature at the sediment was measured over time in a single natural and experimental cryoconite hole.

Methods

Site description and sampling

The experiment was conducted on Canada Glacier, a polar alpine glacier in the Taylor Valley, one of the McMurdo Dry Valleys in Antarctica. We selected Canada Glacier primarily for the presence of extensive information on its cryoconite holes (e.g., Christner et al. 2003; Porazinska et al. 2004; Tranter et al. 2004; Bagshaw et al. 2007; Foreman et al. 2007; Fountain et al. 2008; Tedesco et al. 2013; Telling et al. 2014) but also for its accessibility by foot from the Lake Hoare field camp.

Cryoconite holes

To construct the experimental holes, we collected sediments from the surface of the glacier (“supraglacial sediments”) from the lower Canada Glacier (77.62632°S, 162.94595°E) (Fig. 1a). Before aliquoting, sediments stored at − 20 °C were allowed to thaw at 4 °C for 24 h in the field lab, then homogenized in a sterilized basin. Subsamples of 20 g (wet) (equivalent to ~ 2-mm-thick layer of sediment) were transferred to sterile plastic bags (Whirl–Pak, Nasco, WI, USA) and stored at 4 °C overnight.

Fig. 1
figure1

The process of making experimental cryoconite holes. a Supraglacial sediments used as input sediments to make experimental cryoconite holes. Backpack (70 L) in the foreground for scale. b Input sediments placed in drilled depression on ice. c Melted glacial ice added to sediments. d Temperature probes, with wires wrapped around bamboo, mounted on masts and inserted into the ice to record temperature within experimental cryoconite holes after melt-in. e Melted experimental cryoconite holes being sampled with a serological pipet through a hole drilled in the ice lid

On 24 November, 2016, we drilled 25 cylindrical holes less than 5 cm deep and 8 cm in diameter using a sterilized drill bit (Forstner bit). The sediments were spread into the holes in a layer about 2 mm thick (Fig. 1b). To anchor (freeze) the sediments in place, 10 ml of melted glacial water was added to each hole the following day (Fig. 1c). One subsample was reserved to characterize the community composition and structure of the initial community (“input sediment”), after treatment identical to the other samples.

Within five days’ time, all sediment patches melted into the ice as expected from natural hole formation. We selected 7 out of 25 holes for sampling on 23 December (~1 month later) by gently drilling through the ice lid with an 18 v drill bit cleaned between samples by drilling into bare ice. Six out of the 7 holes contained water, and only these were sampled. We collected 3 × 1-ml sediment subsamples from each hole with a sterile 25-ml serological pipet (Fig. 1e). The three subsamples were pooled in a sterile plastic bag (Whirl–Pak: Nasco, WI, USA) and stored at − 20 °C for up to three weeks before DNA extraction.

On 16 January (~ 2 months later), we sampled 8 out of initial 25 holes. Because all but one were entirely frozen, we drilled by hand into each hole with an ice auger (Kovacs, OR, USA) (cleaned in ice between samples) until the sediments evacuated to the surface, then collected sediments with an ethanol-sterilized steel scoop into sterile plastic bags (Whirl–Pak, Nasco, WI, USA). Among the 8 sampled holes, 4 had been sampled in December. All collected sediment samples were stored at − 20 °C up to three weeks before DNA extraction.

To assess thermal characteristics of the experimental holes, we installed a temperature thermistor in one hole on 27 November, 2016. The thermistor was secured to a stainless steel mast covered white Teflon tape (to minimize solar heating), and positioned 5 cm beneath the ice surface to measure the sediment temperature as it migrated downward (Fig. 1d). The instrumented hole was never sampled for microbial community structure to avoid disturbance. We also inserted a thermistor into a nearby natural cryoconite hole in the same configuration with the probe positioned at the sediment surface. The natural cryoconite hole was approximately 20 cm diameter. To measure the background ice temperature of the glacier, a thermistor was installed one meter below the surface, and for local air temperature one was mounted in a radiation shield one meter above the ice.

As a comparison to the experimental holes, we selected and sampled 8 natural cryoconite holes (14–28 cm diameter) on Canada Glacier to characterize their communities. While still frozen, these holes were sampled on 7 and 8 November, 2016 using a 10-cm diameter ice corer (SIPRE corer). Extracted cores were placed in sterile bags, and stored at − 20 °C for up to four weeks before processing. In the Crary Laboratory at McMurdo Station, the sediment puck (the disc-shared sediment portion of the core) from each core was separated from the overlaying water column and washed with Milli-Q filtered water to remove the outer layer, which was likely cross-contaminated by the corer, then allowed to melt in an acid-washed container at 4 °C for 24 h before being homogenized and subsampled for DNA extraction.

We also sampled other supraglacial sediments from different locations (8 in total) on Canada Glacier on 10 November, 2016 as reference points. An ethanol-sterilized steel scoop was used to transfer sediments into a sterile plastic bag (Whirl–Pak, Nasco, WI, USA). Sediments were stored at − 20 °C up to one month until DNA extraction.

DNA sequencing and data processing

Approximately 0.3 g (wet weight) per each collected sediment sample was used for DNA extraction using a PowerSoil DNA Isolation Kit (MoBio Inc., CA, USA) following the manufacturer’s instructions. Extracted genomic DNA was amplified in triplicate using 16S (515f-806r primers, Caporaso et al. 2012) and 18S (1391f-EukBr primers, Amaral-Zettler et al. 2009; Caporaso et al. 2012) SSU ribosomal gene markers. Amplified DNA was pooled and normalized to equimolar concentrations using SequalPrep Normalization Plate Kit (Invitrogen Corp., CA, USA), and sequenced on two lanes with Illumina MiSeq V2 (Illumina Inc., CA, USA) with 2 × 250 bp chemistry at the BioFrontiers Sequencing Core Facility at the University of Colorado at Boulder.

QIIME v1.9.1 (Caporaso et al. 2010) was used to de-multiplex and quality filter the raw reads. Paired-end sequences were joined with VSEARCH (Rognes et al. 2016) and clustered into operational taxonomic units (OTUs) at 97% similarity using UCLUST (Edgar 2010). Taxonomy was assigned using QIIME’s parallel_assign_taxonomy_blast.py script with the SILVA 128 Ref NR99 database (Quast et al. 2013). Based on this classification, mitochondrial and chloroplast OTUs were removed from the bacterial OTU table, and bacterial OTUs were removed from the eukaryotic OTU table. OTUs that made up at least 1% of the extraction blank sequences and were at least 1% of the samples on average were discarded as likely lab contaminants, with two bacterial exceptions (two Burkholderiales phylotypes) that were also dominant members of the cryoconite assemblage sequences and closely related to other sequences from polar environments. Singletons were removed. Bacterial OTU tables were rarefied to 8700 reads, and eukaryotic OTU tables were rarefied to 5280 reads.

Analyses

We used a linear mixed model implemented with function ‘lme’ in package ‘nlme’ (Pinheiro et al. 2018) in R v 3.5.2 (R Core Team 2018) to compare the total richness of phylotypes (OTUs) between experimental holes sampled in December (n = 6) and those sampled in January (n = 8) with a random intercept term to account for the holes that were resampled at both time points (n = 4) subset of December and January. We used a one-sample t test to test the null hypothesis that the mean phylotype richness in December (n = 6) did not differ from the phylotype richness of the input sediment (n = 1). Normal distributions were used for these tests because the data for the most part did not deviate significantly from assumptions of normality or show significant heteroskedasticity (Online Resource 1).

We used a redundancy analysis (RDA) on the Hellinger distances of unrarefied data (McMurdie and Holmes 2014), and a permutational ANOVA-like test with 999 permutations to ask whether days since establishment of experimental cryoconite holes were a significant correlate of community turnover. We furthermore used a permutational multivariate analysis of variance (PERMANOVA) as implemented by the function ‘adonis’ in package ‘vegan’ (Oksanen et al. 2018) with 999 permutations to compare experimental cryoconite holes sampled in different months, along with other supraglacial sediments and natural cryoconite holes. We conducted post hoc pairwise comparisons using function ‘pariwise.perm.manova’ from package ‘RVAideMemoire’ (Hervé 2018).

Finally, we compared the relative abundance of the dominant bacterial and eukaryotic phylotypes from samples collected in December to those collected in January using linear mixed-effects models and to the input sediment using a one-sample t test. Because the data generally did not deviate from assumptions of normality nor showed heteroskedacity (Online Resource 1), all tests were performed using normal distributions. However, log-transformed relative abundance was used for one eukaryotic phylotype to meet normality assumptions because negative binomial models did not converge (Online Resource 1).

Results

Visual observations

The austral summer of 2016–2017 was snowy. Although the study site was initially snow-free, subsequent snow events required the site to be cleared of snow before experimental holes were installed. Snow over the ice, in this environment, reduces solar radiation into the ice and subsurface warming such that subsurface melt is absent. Fortunately, surface melt was observed at the experimental holes as early as three days after sediment placement (Fig. 2a). By the first sampling date one month later, the experimental cryoconite holes were visibly indistinguishable from natural holes (Fig. 2b, c), with a solid ice lid partially covered with a wind-crust layer of snow, and liquid water inside. By mid-January (approximately 2 months), much of the snow crust persisted, but all sampled holes were frozen.

Fig. 2
figure2

Images illustrating the physical status of cryoconite holes throughout the season. a Initial melt two days after placement. b Approximately 1-month-old experimental cryoconite hole with an ice lid as observed on 23 December, 2016. c A nearby natural cryoconite hole of approximately same size was visually indistinguishable from the experimental hole

Bacteria

In comparison to the input sediment, bacterial phylotype richness (Fig. 3a) declined by the December sampling (one-sample t test, t5 = − 8, p < 0.001) and remained low through the January sampling (one-sample t test, t7 = − 10.43, p < 0.001), with no difference between sampling dates (LME model Jan. to Dec., t3 = − 2.03, p = 0.14, full model results in Online Resource 3).

Fig. 3
figure3

Phylotype richness of a bacteria and b eukaryotes in: Input sediment (n = 1) denoted as dashed line; approximately 1-month-old (n = 6) and 2-month-old (n = 8) experimental cryoconite holes denoted as “Dec” and “Jan”, respectively; supraglacial sediments (n = 8) denoted “Sed;” and natural cryoconite holes (n = 8), denoted “Nat.” Samples from the same experimental hole in December and January are indicated with dotted lines connecting points

The community structure within the experimental cryoconite holes underwent change from the input sediment to December samples, and again by the January samples (RDA, F1 = 2.93, p = 0.001, 18% of variance constrained, full model results in Online Resource 4) (Fig. 4a). Although the bacterial assemblages changed between time points, they remained distinct from natural cryoconite holes and from other supraglacial sediments (PERMANOVA, F3 = 5.9, p = 0.001; pairwise comparisons in Online Resource 2).

Fig. 4
figure4

Redundancy analysis of Hellinger distances of experimental cryoconite hole community structure using the days since they were established for a bacteria and b eukaryotes. Days since establishment (Time: RDA1) explained 18.3% of the variation for bacterial communities and 14.4% for eukaryotic communities

The relative abundance of the dominant Nostoc sp. (Cyanobacteria) over the next most abundant phylotype, Chamaesiphon sp. (Cyanobacteria) increased (Fig. 5a–c) from the input sediment to the December sampling (one-sample t test, t5 = 3.5, p = 0.003), and the Nostoc sp. retained its dominance into January (LME model, t3 = − 2.03, p = 0.14, full model results in Online Resource 3). A Flavobacterium sp. (Bacteroidetes) showed a similar pattern, increasing in relative abundance from initially rare to become one of the dominant phylotypes (one-sample t test of input to Dec., t5 = 2.4, p = 0.03, LME model Dec. to Jan. t3 = − 2.03, p = 0.14, full model results in Online Resource 3) (Fig. 5a–c).

Fig. 5
figure5

Relative abundance of ten most dominant bacterial phylotypes from a input sediments; b experimental holes sampled in December; c experimental holes sampled in January; d other supraglacial sediments; and e naturally occurring cryoconite holes. Error bars are bootstrapped 95% confidence intervals

Eukaryotes

Similar to bacteria, the richness of eukaryotic phylotypes in experimental cryoconite holes declined from the input sediment to December sampling (one-sample t test, t5 = 5.9, df = 5, p = 0.002) (Fig. 3b), and continued to decline over time (LME model Jan. to Dec., t3 = − 4.4, p = 0.02, full model results in Online Resource 3).

The assemblages of microbial eukaryotes also changed significantly over time (RDA, F1 = 2.20, p = 0.005, 14.5% of variance constrained, full model results in Online Resource 4) (Fig. 4b). Similarly to bacteria, the eukaryotic assemblages in experimental holes were different from those in natural cryoconite holes and from supraglacial sediments (PERMANOVA, F3 = 5.6, p = 0.001, pairwise comparisons in Online Resource 2).

Eukaryotic assemblages were dominated by a single phylotype of algae, Macrochloris sp., whose relative abundance increased from < 20% of the sequences in the input sediment to up to ~ 30% in December (one-sample t test start to Dec., t5 = − 3.4, p = 0.019) and January (LME model Dec-Jan comparison, t3 = 1.28, p = 0.29, full model results in Online Resource 3). In contrast, one of initially dominant bdelloid rotifers, Rotaria sp., declined in proportion over time (one-sample t test, t5 = − 5.2, p = 0.003) and from December to January (LME model, t3 = − 6.5, p = 0.008, full model results in Online Resource 3). A phylotype most closely matching the tardigrade Acutuncus antarcticus also initially declined (one-sample t test, t5 = − 4.6, p = 0.006), but then rebounded from December to January (LME model: t3 = 1.4, p = 0.025, full model in Online Resource 3) (Fig. 6a–c).

Fig. 6
figure6

Relative abundance of ten most dominant eukaryotic phylotypes from a input sediments; b experimental holes sampled in December; c experimental holes sampled in January; d other supraglacial sediments; and e naturally occurring cryoconite holes. Error bars are bootstrapped 95% confidence intervals

Temperature

Temperature in the experimental and natural cryoconite holes initially differed likely due to their difference in depth (Fig. 7). In December, temperatures in the natural holes were well below freezing, as expected, with diurnal fluctuations. Over this period, the ice temperature was warming and experimental holes were melting into the ice. By late December, the temperatures in the natural holes were similar to air temperatures, and warmer than air temperatures by mid-January. However, they continued to freeze daily. Temperatures in the experimental holes were warmer than in the natural holes throughout the study period until mid-January when temperatures converged, most likely due to similar depths of the probes. The probe in the natural hole was measured to be 33 cm deep at that time, and the probe in the experimental hole 32 cm deep. Sustained melting over a period of weeks did not occur, as can be seen by the diurnal fluctuations of temperature which always cooled to below freezing after initially warming to melting temperatures in mid-December.

Fig. 7
figure7

Temperature profiles over the course of the season: inside the experimental cryoconite hole (Texp: purple line), natural hole (Tnat: teal line), glacial ice at 1 m depth (Tglcr: black line) and air temperature (Tair: dotted black line). Red vertical dashed line indicates convergence of temperatures inside the experimental and natural cryoconite holes

Discussion

Our results build on past work experimentally creating cryoconite holes (MacDonell and Fitzsimons 2008) by characterizing their subsequent biological and physical dynamics over a growing season. The temperature profiles of experimental cryoconite holes fluctuated daily as they melted into the ice, forming meltwater under an ice lid. Hole appearance was already indistinguishable from natural holes by mid-season. Moreover, by the end of the season, they achieved the same temperature profile as natural holes. Changes in the microbial assemblages over that time period were indicative of microbial species turnover and growth and confirmed our hypothesis that we would be able to detect changes in community structure within a single growing season. This finding lays the foundation for the use of cryoconite holes as experimental mesocosms for studying microbial community dynamics.

Biologically, experimental cryoconite holes developed their own signatures and dynamics that differed from all sediment types collected in this study. The richness of both the bacterial and eukaryotic phylotypes in experimental holes was lower than that of the input sediment and eukaryotic richness continued to decrease over time. The reduction in richness following division of the initial sediment into smaller “islands” of sediment in individual experimental holes is consistent with patterns of island biogeography theory (MacArthur and Wilson 1967), one of the few consistent relationships in ecology, and one which cryoconite holes follow at this site and globally (Darcy et al. 2018). Smaller islands may reduce species richness through stochastic extinctions (MacArthur and Wilson 1967) or reduced habitat variability and thus opportunity to partition ecological niches (Hortal et al. 2009). Such niche partitioning is fundamental to stable coexistence of species (Chesson 2000). Although our current data do not allow tests of the mechanisms driving island biogeography theory, it supports the use of experimental cryoconite holes in future tests of these hypotheses.

An alternative but possible explanation for the decline in richness within experimental holes could be that some taxa in sediments are not active, but perhaps represent relic DNA atmospherically deposited by high winds that move sediment through the valley ( Lancaster et al. 2002; Nylen et al. 2004; Šabacká et al. 2012; Diaz et al. 2018). Higher rates of microbial activity may be supported within cryoconite holes than on the glacial surface by the more stable temperatures we measured, buffered by the availability of liquid water for microbes, and lower light stress (Bagshaw et al. 2016). Relic DNA would therefore persist longer at the surface than in more active melted sediments, which would be consistent with previous findings of distinct phylotypes in sediments and surface ice of cryoconite holes (Sommers et al. 2019). Experimental cryoconite holes could be used in future work to test the rate at which known quantities of dead organisms’ DNA is degraded in this environment.

In addition to changes in richness, both bacterial and eukaryotic assemblages underwent changes in composition throughout the growing season. For example, the relative abundance of a phylotype most closely matching Hymenobacter antarcticus (Bacteroidetes: accession GQ45800.1, Klassen and Foght 2011) increased to become the third most abundant phylotype by January in experimental holes. In contrast, the same phylotype was rare in natural holes and in other supraglacial sediments. Other members of this genus have been isolated in Antarctica from red snow (Kojima et al. 2016), and glacial water (Marizcurrena et al. 2017), perhaps indicating it can take advantage of shifting environmental conditions. Similarly, the relative abundance of the dominant cyanobacteria, Chamaesiphon sp. and Nostoc sp., and the dominant eukaryotic phototroph, Macrochloris sp., increased. Cyanobacteria from the genus Nostoc are commonly found in cryoconite on polar glaciers (Segawa et al. 2017). These communities did not match those of natural cryoconite holes sampled by the end of the season, and it is possible they never will due to differences in source material, as the sources and ages of the natural holes are unknown. It is also possible that these communities could be undergoing ecological selection by new environmental conditions (Vellend 2010) and that it may take more than one season for communities to mature to a stable structure, if one exists. Stable structures of bacterial communities in cryoconite holes have been observed in Greenland, despite the fact that those holes lacked ice lids and could receive continual input from the atmosphere and from supraglacial melt (Musilova et al. 2015). By contrast, bacterial communities of alpine cryoconite holes vary throughout seasons in the Alps (Franzetti et al. 2017; Pittino et al. 2018), although comparison of these results with Dry Valley cryoconite holes should be made cautiously, as alpine and Arctic cryoconite holes are typically unlidded and can have substantial hydrologic connectivity relative to cryoconite holes in the Dry Valleys (Mueller and Pollard 2004).

The dominance of primary producers could be indicative of favorable conditions for their growth, such as availability of water and photosynthetically active radiation (Bagshaw et al. 2016). It could also indicate a release from potential grazers such as rotifers and tardigrades, which decreased in their relative abundance. Amplicon sequencing data are compositional, meaning that it represents relative abundances rather than absolute abundances, so a faster increase in producers would necessarily result in an apparent reduction in microfauna, but our data cannot distinguish between rapid growth of one organism and actual decline of another. Sequencing data are furthermore influenced by differing gene copy numbers between organisms (Větrovský and Baldrian 2013), meaning that abundances of sequences should not be interpreted as the abundances of organisms. However, when the same set of samples is followed through time, as they are here, the changes in the abundance of organisms relative to one another can be used to infer changes in the relative abundance of those organisms that indicate change through time.

Despite the fact that the 2016–2017 was snowier than past years, the experimental holes exhibited shifting microbial communities indicating biological activity. Their morphology was much more stable than cryoconite holes monitored (Takeuchi et al. 2018) or manipulated (Cook et al. 2016) in Greenland, as is typical of natural cryoconite holes in the McMurdo Dry Valleys of Antarctica (Fountain et al. 2004). Also in contrast to Arctic and alpine cryoconite holes, an ice lid formed on experimental holes, consistent with natural holes in the Dry Valleys (Mueller et al. 2001). Neither experimental nor natural holes showed the same seasonal melt-development pattern that has been measured during several previous, largely snow-free summers on the same glacier (Fountain et al. 2004, 2008). Although we cleared some of the snow from the experimental study area, the remaining snow trapped in the rough ice surface greatly reduced solar heating of the ice, a critical factor in this environment because summer air temperatures are typically a few degrees below freezing (Zamora 2018; Hoffman et al. 2008). Furthermore, the ice surrounding the study site was not cleared away, creating a heat sink that further cooled the study site. This likely contributed to the temperature consistently dipping below freezing daily for both natural and experimental holes, even in January, in contrast to the previous, snow-free years (Zamora 2018).

In conclusion, within a single season, our experimental cryoconite holes mimicked the conditions in naturally formed cryoconite holes. Over that same time period, the assemblages of bacteria and microbial eukaryotes underwent compositional changes indicative of natural community dynamics, activity, and growth. Together, these data provide support for the use of experimental cryoconite holes as mesocosms to test hypotheses from community ecology theory (e.g., community assembly). Future work could, for instance, manipulate input to cryoconite holes to test whether the order in which they arrive (priority effects) is important (Fukami 2015), or whether the relationship between taxonomic composition and ecosystem functioning differs between communities of microscopic organisms and larger organisms (Nemergut et al. 2013).

References

  1. Amaral-Zettler LA, McCliment EA, Ducklow HW, Huse SM (2009) A method for studying protistan diversity using massively parallel sequencing of V9 hypervariable regions of small-subunit ribosomal RNA genes. PLoS ONE 4:e6372. https://doi.org/10.1371/journal.pone.0006372

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  2. Anesio AM, Hodson AJ, Fritz A, Psenner R, Sattler B (2009) High microbial activity on glaciers: importance to the global carbon cycle. Glob Chang Biol 15:955–960. https://doi.org/10.1111/j.1365-2486.2008.01758.x

    Article  Google Scholar 

  3. Bagshaw EA, Tranter M, Fountain AG, Welch KA, Basagic H, Lyons WB (2007) Biogeochemical evolution of cryoconite holes on Canada glacier, Taylor Valley, Antarctica. J Geophys Res Biogeosci 112:G4. https://doi.org/10.1029/2007JG000442

    CAS  Article  Google Scholar 

  4. Bagshaw EA, Tranter M, Fountain AG, Welch K, Basagic HJ, Lyons WB (2013) Do cryoconite holes have the potential to be significant sources of C, N, and P to downstreamdepauperate ecosystems of Taylor Valley, Antarctica? Arct Antarct Alp Res 45:440–454. https://doi.org/10.1657/1938-4246-45.4.440

    Article  Google Scholar 

  5. Bagshaw EA, Wadham JL, Tranter M, Perkins R, Morgan A, Williamson CJ, Fountain AG, Fitzsimons S, Dubnick A (2016) Response of Antarctic cryoconite microbial communities to light. FEMS Microb Ecol 92:fiw076. https://doi.org/10.1093/femsec/fiw076

    CAS  Article  Google Scholar 

  6. Cameron KA, Hodson AJ, Osborn AM (2012) Structure and diversity of bacterial, eukaryotic and archaeal communities in glacial cryoconite holes from the Arctic and the Antarctic. FEMS Microbiol Ecol 82:254–267. https://doi.org/10.1111/j.1574-6941.2011.01277.x

    CAS  Article  PubMed  Google Scholar 

  7. Caporaso JG, Kuczynski J, Stombaugh J, Bittinger K, Bushman FD, Costello EK, Fierer N, Peña AG, Goodrich JK, Gordon JI, Huttley GA, Kelley ST, Knights D, Koenig JE, Ley RE, Lozupone CA, McDonald D, Muegge BD, Pirrung M, Reeder J, Sevinsky JR, Turnbaugh PJ, Walters WA, Widmann J, Yatsunenko T, Zaneveld J, Knight R (2010) QIIME allows analysis of high-throughput community sequencing data. Nat Meth 7:335–336. https://doi.org/10.1038/nmeth.f.303

    CAS  Article  Google Scholar 

  8. Caporaso JG, Lauber CL, Walters WA, Berg-Lyons D, Huntley J, Fierer N, Owens SM, Betley J, Fraser L, Bauer M, Gormley N, Gilbert JA, Smith G, Knight R (2012) Ultra-high-throughput microbial community analysis on the Illumina HiSeq and MiSeq platforms. ISME J 6:1621–1624. https://doi.org/10.1038/ismej.2012.8

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  9. Chesson P (2000) Mechanisms of maintenance of species diversity. Ann Rev Ecol Syst 31:343–366. https://doi.org/10.1146/annurev.ecolsys.31.1.343

    Article  Google Scholar 

  10. Christner BC, Kvitko BH, Reeve JN (2003) Molecular identification of bacteria and eukarya inhabiting an Antarctic cryoconite hole. Extremophiles 1:177–183. https://doi.org/10.1007/s00792-002-0309-0

    CAS  Article  Google Scholar 

  11. Cook J, Hodson A, Telling J, Anesio A, Irvine-Fynn T, Bellas C (2010) The mass–area relationship within cryoconite holes and its implications for primary production. Ann Glaciol 51:106–110. https://doi.org/10.3189/172756411795932038

    CAS  Article  Google Scholar 

  12. Cook JM, Hodson AJ, Anesio AM, Hanna E, Yallop M, Stibal M, Telling J, Huybrechts P (2012) An improved estimate of microbially mediated carbon fluxes from the Greenland ice sheet. J Glaciol 58:1098–1108. https://doi.org/10.3189/2012JoG12J001

    Article  Google Scholar 

  13. Cook JM, Edwards A, Bulling M, Mur LA, Cook S, Gokul JK, Cameron KA, Sweet M, Irvine-Fynn TD (2016) Metabolome-mediated biocryomorphic evolution promotes carbon fixation in Greenlandic cryoconite holes. Env Microbiol 18:4674–4686. https://doi.org/10.1111/1462-2920.13349

    CAS  Article  Google Scholar 

  14. Darcy JL, Gendron EMS, Sommers P, Porazinska DL, Schmidt SK (2018) Island biogeography of cryoconite hole bacteria in Antarctica’s Taylor Valley and around the world. Front Ecol Evol 6:180. https://doi.org/10.3389/fevo.2018.00180

    Article  Google Scholar 

  15. Diaz MA, Adams BJ, Welch KA, Welch SA, Opiyo SO, Khan AL, Lyons WB et al (2018) Aeolian material transport and its role in landscape connectivity in the McMurdo Dry Valleys, Antarctica. J Geophys Res Earth Surf 123:3323–3337. https://doi.org/10.1029/2017JF004589

    Article  Google Scholar 

  16. Edgar RC (2010) Search and clustering orders of magnitude faster than BLAST. Bioinform 26:2460–2461. https://doi.org/10.1093/bioinformatics/btq461

    CAS  Article  Google Scholar 

  17. Foreman CM, Sattler B, Mikucki JA, Porazinska DL, Priscu JC (2007) Metabolic activity and diversity of cryoconites in the Taylor Valley. Antarctica. J Geophys Res 112:G4. https://doi.org/10.1029/2006JG000358

    CAS  Article  Google Scholar 

  18. Fountain AG, Nylen TH, Tranter M, Bagshaw E (2008) Temporal variations in physical and chemical features of cryoconite holes on Canada Glacier, McMurdo Dry Valleys, Antarctica. J Geophys Res Biogeosci 113:92.https://doi.org/10.1029/2007JG000430

    Article  Google Scholar 

  19. Fountain AG, Tranter M, Nylen TH, Lewis KJ, Mueller DR (2004) Evolution of cryoconite holes and their contribution to meltwater runoff from glaciers in the McMurdo Dry Valleys, Antarctica. J Glaciol 50:35–45. https://doi.org/10.3189/172756504781830314

    Article  Google Scholar 

  20. Franzetti A, Navarra F, Tagliaferri I, Gandolfi I, Bestetti G, Minora U, Azzoni RS, Diolaiuti G, Smiraglia C, Ambrosini R (2017) Temporal variability of bacterial communities in cryoconite on an Alpine glacier. Environ Microbiol Rep 9(71):78. https://doi.org/10.1111/1758-2229.12499

    CAS  Article  Google Scholar 

  21. Fukami T (2015) (2015) Historical contingency in community assembly: Integrating niches, species pools, and priority effects. Ann Rev Ecol Evol Syst 46:1–23. https://doi.org/10.1146/annurev-ecolsys-110411-160340

    Article  Google Scholar 

  22. Gribbon PW (1979) Cryoconite holes on Sermikavsak, west Greenland. J Glaciol 22:177–181. https://doi.org/10.3189/S0022143000014167

    Article  Google Scholar 

  23. Hervé M (2018) RVAideMemoire: testing and plotting procedures for biostatistics. R package version 0.9–69–3. https://CRAN.R-project.org/package=RVAideMemoire

  24. Hoffman MJ, Fountain AG, Liston GE (2008) Surface energy balance and melt thresholds over 11 years at Taylor Glacier, Antartica. J Geophys Res Earth Surf 113:F04014. https://doi.org/10.1029/2008JF001029

    Article  Google Scholar 

  25. Hortal J, Triantis KA, Meiri S, Thébault E, Sfenthourakis S (2009) Island species richness increases with habitat diversity. Am Nat 174:E205–271. https://doi.org/10.1086/645085

    Article  PubMed  Google Scholar 

  26. Klassen JL, Foght JM (2011) Characterization of Hymenobacter isolates from Victoria Upper Glacier, Antarctica reveals five new species and substantial non-vertical evolution within this genus. Extremophiles 15:45–57. https://doi.org/10.1007/s00792-010-0336-1

    Article  PubMed  Google Scholar 

  27. Kojima H, Watanabe M, Tokizawa R, Shinohara A, Manabu F (2016) Hymenobacter nivis sp. nov., isolated from red snow in Antarctica. Int J Syst Evol Microbiol 66:4821–4825. https://doi.org/10.1099/ijsem.0.001435

    CAS  Article  PubMed  Google Scholar 

  28. Lancaster N (2002) Flux of eolian sediment in the McMurdo Dry Valleys, Antarctica: a preliminary assessment. Arct Antarct Alp Res 34:318–323. https://doi.org/10.1080/15230430.2002.12003500

    Article  Google Scholar 

  29. MacArthur RH, Wilson EO (1967) The theory of island biogeography. Princeton University Press, New Jersey

    Google Scholar 

  30. MacDonell S, Fitzsimons S (2008) The formation and hydrological significance of cryoconite holes. Prog Phys Geog 32:595–610. https://doi.org/10.1177/0309133308101382

    Article  Google Scholar 

  31. Marizcurrena JJ, Morel MA, Braña V, Morales D, Martinez-López W, Castro-Sowinski S (2017) Searching for novel photolyases in UVC-resistant Antarctic bacteria. Extremophiles 21:409–418. https://doi.org/10.1007/s00792-016-0914-y

    CAS  Article  PubMed  Google Scholar 

  32. McMurdie PJ, Holmes S (2014) Waste not, want not: Why rarefying microbiome data is inadmissible. PLoS Comput Biol 10:e1003531. https://doi.org/10.1371/journal.pcbi.1003531

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  33. Mueller DR, Pollard WH (2004) Gradient analysis of cryoconite ecosystems from two polar glaciers. Polar Biol 27:66–74. https://doi.org/10.1007/s00300-003-0580-2

    Article  Google Scholar 

  34. Mueller DR, Vincent WF, Pollard WH, Fritsen CH (2001) Glacial cryoconite ecosystems: a bipolar comparison of algal communities and habitats. Nova Hedwig Beih 123:173–198

    Google Scholar 

  35. Musilova M, Tranter M, Bennett SA, Wadham J, Anesio AM (2015) Stable microbial community composition on the Greenland Ice Sheet. Front Microbiol 6:193. https://doi.org/10.3389/fmicb.2015.00193

    Article  PubMed  PubMed Central  Google Scholar 

  36. Nemergut DR, Schmidt SK, Fukami T, O’Neill SP, Bilinski TM, Stanish LF, Knelman JE, Darcy JL, Lynch RC, Wickey P, Ferrenberg S (2013) Patterns and processes of microbial community assembly. Microbiol Molec Biol Rev 77:342–356. https://doi.org/10.1128/MMBR.00051-12

    Article  Google Scholar 

  37. Nordenskjöld NE (1875) Cryoconite found 1870, July 19th–25th, on the inland ice, east of Auleitsivik Fjord, Disco Bay, Greenland. Geolog Mag 2:157–162

    Google Scholar 

  38. Nylen TH, Fountain AG, Doran PT (2004) Climatology of katabatic winds in the McMurdo Dry Valleys, southern Victoria Land. Antarctica J Geophys Res 109:D3. https://doi.org/10.1029/2003JD003937

    Article  Google Scholar 

  39. Oksanen J, Guillaume Blanchet F, Friendly M, Kindt R, Legendre P, McGlinn D, Minchin PR, O’Hara RB, Simpson GL, Solymos P, Stevens MHH, Szoecs E, Wagner H (2018) vegan: Community Ecology Package. R package version 2.5-1. https://CRAN.R-project.org/package=vegan

  40. Pinheiro J, Bates D, DebRoy S, Sarkar D, R Core Team (2018) nlme: Linear and Nonlinear Mixed Effects Models. R package version 3.1–137, https://CRAN.R-project.org/package=nlme.

  41. Pittino F, Maglio M, Gandolfi I, Azzoni RS, Diolaiuti G, Ambrosini R, Franzetti A (2018) Bacterial communities of cryoconite holes of a temperate alpine glacier show both seasonal trends and year-to-year variability. Ann Glaciol. https://doi.org/10.1017/aog.2018.16s

    Article  Google Scholar 

  42. Porazinska DL, Fountain AG, Nylen TH, Tranter M, Virginia RA, Wall DH (2004) The biodiversity and biogeochemistry of cryoconite holes from McMurdo Dry Valley glaciers, Antarctica. Arct Antarct Alp Res 36:84–91. https://doi.org/10.1657/1523-0430(2004)036[0084:TBABOC]2.0.CO;2

    Article  Google Scholar 

  43. Quast C, Pruesse E, Yilmaz P, Gerken J, Schweer Yarza P, Peplies J, Glöckner FO (2013) The SILVA ribosomal RNA gene database project: Improved data processing and web-based tools. Nucl Acids Res 41:D590–D596. https://doi.org/10.1093/nar/gks1219

    CAS  Article  PubMed  Google Scholar 

  44. R Core Team (2018) R: A language and environment for statistical computing. v3.3.2 https://www.R-project.org/ Vienna, Austria. R Foundation for Statistical Computing.

  45. Rognes T, Flouri T, Nichols B, Quince C, Mahé F (2016) VSEARCH: A versatile open source tool for metagenomics. PeerJ 4:e2584. https://doi.org/10.7717/peerj.2584

    Article  PubMed  PubMed Central  Google Scholar 

  46. Šabacká M, Priscu JC, Basagic HJ, Fountain AG, Wall DH, Virginia RA, Greenwood MC (2012) Aeolian flux of biotic and abiotic material in Taylor Valley, Antarctica. Geomorphol 155:102–111. https://doi.org/10.1016/j.geomorph.2011.12.009

    Article  Google Scholar 

  47. Segawa T, Yonezawa T, Edwards A, Akiyoshi A, Tanaka S, Uetake J, Irvine-Fynn T, Fukui K, Li Z, Takeuchi N (2017) Biogeography of cryoconite forming cyanobacteria on polar and Asian glaciers. J Biogeog 44:2859–2861. https://doi.org/10.1111/jbi.13089

    Article  Google Scholar 

  48. Sommers P, Darcy JL, Gendron EM, Stanish LF, Bagshaw EA, Porazinska DL, Schmidt SK (2018) Diversity patterns of microbial eukaryotes mirror those of bacteria in Antarctic cryoconite holes. FEMS Microbiol Ecol 94:167. https://doi.org/10.1093/femsec/fix167

    CAS  Article  Google Scholar 

  49. Sommers P, Darcy JL, Porazinska DL, Gendron EM, Fountain AG, Zamora F, Vincent K, Cawley KM, Solon AJ, Vimercati L, Ryder J, Schmidt SK (2019) Comparison of microbial communities in the sediments and water columns of frozen cryoconite holes in the McMurdo Dry Valleys. Antarctica. Front Microbiol 10:65

    Article  PubMed  Google Scholar 

  50. Stanish LF, Bagshaw EA, McKnight DM, Fountain AG, Tranter M (2013) Environmental factors influencing diatom communities in Antarctic cryoconite holes. Env Res Lett 8:045006. https://doi.org/10.1088/1748-9326/8/4/045006

    Article  Google Scholar 

  51. Takeuchi N, Sakaki R, Uetake J, Nagatsuka N, Shimada R, Niwano M, Aoki T (2018) Temporal variations of cryoconite holes and cryoconite coverage on the ablation ice surface of Qaanaaq Glacier in northwest Greenland. Ann Glaciol 1:10. https://doi.org/10.1017/aog.2018.19

    Article  Google Scholar 

  52. Tedesco M, Foreman CM, Anton J, Steiner N, Schwartzman T (2013) Comparative analysis of morphological, mineralogical and spectral properties of cryoconite in Jakobshavn Isbrae, Greenland, and Canada Glacier, Antarctica. Ann Glaciol 54:147–157. https://doi.org/10.3189/2013AoG63A417

    Article  Google Scholar 

  53. Telling J, Anesio AM, Tranter M, Fountain AG, Nylen T, Hawkings J, Singh VB, Kaur P, Musilova M, Wadham JL (2014) Spring thaw ionic pulses boost nutrient availability and microbial growth in entombed Antarctic Dry Valley cryoconite holes. Front Microbiol 5:694. https://doi.org/10.3389/fmicb.2014.00694

    Article  PubMed  PubMed Central  Google Scholar 

  54. Tranter M, Fountain AG, Fritsen CH, Lyons WB, Priscu JC, Statham PJ, Welch KA (2004) Extreme hydrochemical conditions in natural microcosms entombed within Antarctic ice. Hydrol Proc 18:379–387. https://doi.org/10.1002/hyp.5217

    Article  Google Scholar 

  55. Vellend M (2010) Conceptual synthesis in community ecology. Q Rev Biol 85:183–206. https://doi.org/10.1086/652373

    Article  PubMed  Google Scholar 

  56. Větrovský T, Baldrian P (2013) The variability of the 16S rRNA gene in bacterial genomes and its consequences for bacterial community analyses. PLoS ONE 8:e57923. https://doi.org/10.1371/journal.pone.0057923

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  57. Webster-Brown JG, Hawes I, Jungblut AD, Wood SA, Christenson HK (2015) The effects of entombment on water chemistry and bacterial assemblages in closed cryoconite holes on Antarctic glaciers. FEMS Microbiol Ecol 91:144. https://doi.org/10.1093/femsec/fiv144

    CAS  Article  Google Scholar 

  58. Wharton RA Jr, McKay CP, Simmons GM Jr, Parker BC (1985) Cryoconite holes on glaciers. BioSci 1:499–503. https://doi.org/10.2307/1309818

    Article  Google Scholar 

  59. Zamora F (2018) Measuring and modeling evolution of cryoconite holes in the McMurdo DryValleys, Antarctica. MS Thesis, Portland State University. https://doi.org/10.15760/etd.6590

Download references

Acknowledgements

This work was conceptualized by the late Diana Nemergut, who is greatly missed. The authors would like to thank all the United States Antarctic Program staff who made these logistics feasible, UNAVCO for precision GPS support, and the BioFrontiers Sequencing Facility at the University of Colorado. Thanks also to Roberto Ambrosini, Jun Uetake, and an anonymous reviewer for comments that improved the manuscript. This work was funded by the United States National Science Foundation Polar Programs Awards 1443578 and 1443373.

Author information

Affiliations

Authors

Corresponding author

Correspondence to Pacifica Sommers.

Ethics declarations

Conflict of interest

The authors declare that they have no conflict of interest.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Sommers, P., Porazinska, D.L., Darcy, J.L. et al. Experimental cryoconite holes as mesocosms for studying community ecology. Polar Biol 42, 1973–1984 (2019). https://doi.org/10.1007/s00300-019-02572-7

Download citation

Keywords

  • Cryoconite
  • Antarctic
  • Bacteria
  • Eukaryotes
  • Algae
  • Cyanobacteria