Essay Writing Service

1 Star2 Stars3 Stars4 Stars5 Stars (No Ratings Yet)

An Exploration of Microbial Mat Community Diversity in Response to Water Level Changes

Cyanobacteria in the Antarctic: An Exploration of Microbial Mat Community Diversity in Response to Water Level Changes Throughout the McMurdo Dry Valleys

BIOSG095: Term III – Research Project II

(1) Life Sciences, Diversity and Bioinformatics, Natural History Museum, Cromwell Road, London, SW7 5BD

(2) Center for Biodiversity and Environment Research, University College London, Gower Street, London, WC1E 6BT



The McMurdo Dry Valleys, Southern Victoria Land, Antarctica, are one of the harshest environments found on earth. The most abundant life forms to inhabit this arid desert are cyanobacteria dominated microbial mats, found throughout perennially ice-covered lakes and ponds. Polar lakes and ponds serve as indicators and predictors of the impacts of environmental change, making the continued study of this regions a priority.  For the purpose of this project, DNA extraction was conducted on samples from three ponds and one lake, Lake Canopus, for 16S genetic sequencing. Samples from the three ponds were collected at three different depth levels of the mat, while samples collected from Lake Canopus were collected along the paleo-shore edge. From the 21 samples a total OTU count of 2470300 was achieved with a mean of 117633.333. Cyanobacteria and proteobacteria were the dominant taxa throughout all samples. PCA, ANOSIM, and SIMPER analysis show there to be significant differences in community structures between the lake and pond locations at the genus level. Results also showed significant differences at the genera level within site locations, with the edge of sites varying from the core samples. Due to their consistent phyla level diversities, cyanobacteria in microbial mats are resilient to water level changes. Understanding the drivers of community structure and their response to environmental pressures, such as water level changes, is of utmost importance as the impacts of climate change continue to grow. With the knowledge of how microbial mat community structure is affected by these environmental pressures we may be able to better predict future responses to climatic shifts as well protect and preserve these invaluable ecosystems.



Keywords: Cyanobacteria, Proteobacteria, Antarctica, McMurdo Dry Valleys, ponds, climate change, microbial mat, Principal Component Analysis (PCA), ANOSIM


Antarctica, and other perennially cold environments were long thought to be devoid of life, and if present to host extremely low levels of biodiversity (Miteva, 2008; Tang, et al., 1997). While there are limited recordings of plant and animal life, research over the past decades have shown there to be an abundance of microscopic life (Chown, et al., 2015; Morgan-Kiss, et al., 2006). One of the most prominent forms of life found throughout these harsh conditions are cyanobacteria dominated microbial mats (Jungblut and Hawes, 2017; Vincent and Quesada, 2012).


1.1 Cyanobacteria

Cyanobacteria are oxygenic, photosynthetic, and Gram-negative bacteria (Jungblut and Hawes, 2017; Tang, et al., 1997; Vincent and Quesada, 2012). Despite being originally described in the 18th century as algae, cyanobacteria are now classified under the nomenclature of bacteria (Jungblut and Hawes, 2017; Oren, 2004; Stanier, et al., 1978). Some cyanobacteria are known to be nitrogen-fixing and to produce toxic compounds known as cyanotoxins (Kleinteich,et al., 2014). These bacteria are the primary producers in both polar and alpine regions and they play an important role not only in the food web, but also as the drivers of both carbon and nitrogen cycling (Chrismas, et al., 2015; Jungblut and Hawes 2017).

Studies have shown cyanobacteria to have incredible diversity and to be tolerant of extreme conditions including those found in the polar regions (Jungblut and Hawes 2017; Morgan-Kiss, et al., 2006). Psychrotolerant cyanobacteria dominate the polar regions as they have a wide array of adaptations to thrive in these environments (Morgan-Kiss, et al., 2006; Varin, et al., 2012). Several studies have identified potential links from genes to psychrotolerant properties such as DNA replication and membrane modifications (Chrismas, et al., 2015; Morgan-Kiss, et al.,  2006; Varin, et al., 2012). It is the combination of resistance and resilience which allows cyanobacterial communities to thrive in extreme conditions. In Antarctic benthic environments, cyanobacteria form cohesive mats and biofilms which are the ecological key stone taxa in polar limnetic systems (Hawes and Jungblut, 2016; Jungblut and Hawes, 2017).

1.2 Microbial Mats


Microbial mats are cyanobacterial-based communities of microorganisms (Jungblut and Hawes, 2017; Velázquez, et al., 2016). These mats are highly productive due to their dominance by diatoms and phototropic cyanobacteria (Velázquez, et al., 2016). Believed to be the first oxygenically active organisms, microbial mats are documented as one of the oldest known groups of organisms (Jungblut, et al., 2005). The mats, and their habitat, are thought to be analogues for the biotopes during the Precambrian glaciation events (Jungblut and Hawes, 2017).


Fig. 1 – A vertical microbial mat showing the accrual of horizontally laminated layers. Figure by Hawes and Jungblut (2016).

These mats are vertically stratified (Fig. 1), organized by function, and are home to other motile species including Nostocales and Oscillatoriales (Hawes and Jungblut, 2016; Jungblut, et al., 2005). Work by Hawes and Jungblut (2016) shows that microbial mats from Lake Hoare, McMurdo Dry Valleys, have both slow accumulation and turnover rates, but display stability in community structure over extended periods of time. Previous research has found that mats accrue in annual, horizontal laminations, allowing for time-scale studies to track the interaction between microbial mats and their environment (Hawes, et al., 2001b). Mats are photosynthetically active during the Antarctic summer, and these layers show the mats are able to survive the 3 months of complete darkness during the Antarctic winters (Hawes, et al., 2001b).

Because of their rich communities, microbial mats are the hotspots of biodiversity in perennially cold environments (Jungblut and Hawes, 2017). Due their biomass and metabolic capabilities, microbial mats are thought be the most important systems in Antarctica (Velázquez, et al.,  2016).


1.3 Antarctica

Home to 70% of the freshwater reservoirs on Earth, Antarctica harbors large collections of cyanobacteria and microbial mats in an array of ecosystems (Jungblut, et al., 2018). Throughout the 20th century the average temperature of the Earth’s surface has increased by 0.6°C per decade, and has continued to climb to an increase of 0.19°C per decade since 1979 (Doran, 2002). This increase in global temperatures has the strongest impacts in the polar regions, with areas in the Antarctic Peninsula warming an average 0.55°C per decade, and >14,000 km2 of the Antarctic ice shelves having collapsed (Verleyen, 2010). However, climate change is having other, less well studied impacts in areas such as the McMurdo Dry Valleys where there has been a recorded drop in average temperature by 0.7°C (Doran, et al.,  2002; Verleyen, et al., 2010). Studying the dramatic shifts in the Antarctic, particularly in areas such as the McMurdo Dry Valleys where there is limited human activity, allows us to create better predictive models of the effects of climate change on a global scale.

1.4 McMurdo Dry Valleys


The McMurdo Dry Valleys, Southern Victoria Land, Antarctica (Fig.3 – A, B) are home to some of the last remaining undisrupted bodies of freshwater ecosystems, with their primary biomass being microbial mats (Velázquez, et al., 2016). The average temperature in this desert is ~-17°C, and it receives the equivalent of less than 10mm of water precipitation, in the form of snow, per year (Kong, et al., 2012). The freezing temperatures are primarily controlled by wind exposure (Doran, et al., 2002). This is the largest ice-free region found in Antarctica, covering an approximate 4,000 km2 (Fountain, et al., 2014; Kong, et al., 2012).

Despite being classified as a hyper-arid cold desert, the McMurdo Dry Valleys house a basin of closed lakes which remain in a liquid state year-round (Kong et al., 2012). These closed lakes are often described as “oases” for their ability to support incredibly rich microbial communities (Hawes and Jungblut, 2016). The McMurdo Dry Valleys are also home to several meltwater ponds. Meltwater ponds are common throughout the Antarctic continent, and provide a lens with which to view the microbial diversity and their adaptations to different environmental pressures. These perennially ice-covered limnetic ecosystems are some of the most physically stable and pristine habitats found on earth due to both their weak hydrological cycles and the extremely cold temperatures (Fountain, et al., 2014; Jungblut, et al.,  2015). The limnetic ecosystems found in the McMurdo Dry Valley are covered with a thick layer of ice, approximately 3-6 inches, which protects the microorganisms from direct contact with atmospheric pressures such as wind turbulence (Kong, et al., 2012). Long-term studies on the cooling of the McMurdo Dry Valleys show that lake water levels are decreasing while lake-ice thickness continues to increase resulting in a drop in the primary production of the lakes (Verlyen, et al., 2010).

1.5 Resistance/Resilience


Microbial ecosystems are subject to constant disturbances, and understanding their reactions to these disturbances will help us better understand the pressures that drive community structure. For the purpose of this project we will define disturbances as any event that directly alters the environment or the community (Shade, et al., 2012). In response to disturbances communities can exhibit two different forms of stability: resistance and resilience (Shade, et al., 2012). Resilience is the ability of the community to return to its original state before the disturbance and resistance is the measure of insensitivity of the community to any given disturbance (Shade, et al., 2012). Microorganisms are thought to be resilient due to a combinations of unique features: (i) rapid growth rates; (ii) physiological flexibility; and (iii) rapid evolution (Allison, 2008).

The harsh environment of the McMurdo Dry Valleys exposes its limnetic ecosystems to drastic changes in temperature and light regimes throughout the year (Archer, et al., 2014). One of the biggest threats microbial mats face is the high levels of UV radiation and high energy photosynthetically active radiation (PAR). Over time, cyanobacteria have evolved several techniques to increase and strengthen DNA repair, but these become less effective the lower the temperatures drop (Chrismas, et al., 2016; Vincent, 2007) Layers of ice covering the bodies of water protect the mats by attenuating approximately 95% of the incident light, as well as setting the solar spectrum to blue-green wavelengths (Kong, et al., 2012). Pigment systems then enacted by the mats absorb between 30-50% of the light depending on their depth within and the location of the lake or pond (Hawes, et al., 2001a). A similar pressure faced by microbial mats is the low levels of irradiances due to long periods of cover by snow or ice, which is combated by the use of light-capturing complexes (Jungblut and Hawes, 2017). Impressively, laboratory studies on microbial mats have shown they can resume photosynthesis within minutes of re-thawing, though this time varies between genera (Tashyreva and Elster, 2016). Rapid changes in salinity levels is yet another pressure faced by microbial mats. Resistance techniques include the production of organic osmolyets, and an increase of organic ions (Jungblut and Hawes, 2017).

Microorganisms drive ecosystem processes, but despite the taxonomic breadth of microbial groups, they are sensitive to environmental disturbances and may not be resilient depending on the type and scale of the disturbances (Allison and Martiny, 2008). The level of resilience of Antarctic communities to long term changes has not been tested yet. However, Antarctic cyanobacteria have been found to have evolved several adaptations to these environmental stressors such as membrane modifications, increases in alternative sigma factor (sigma B) genes, cold shock proteins, EPS, and exopolysaccharides (Varin, et al., 2012). We do not yet know the extent to which the impacts of major microbial community changes will have on isolated and fragile such as Antarctica (Allison and Martiny, 2008).





1.6 Project Overview

Most comparative genomic studies in Antarctica to date have been conducted on the larger lakes due to both their stability as well as their overall influence on the terrestrial systems surrounding them (Archer, et al., 2014). Ponds can be more heavily impacted by climate change both seasonally and globally, and increases studies on ponds will allow us to better understand microbial distribution throughout the Antarctic continent. We have included the sampling of three ponds (L8, L16, and L26), and one lake, Lake Canopus, in our study in order to have a broader scope with which to explore microbial mat community structure and diversity as well as their resilience to water level changes throughout the McMurdo Dry Valleys.

For several years, Lake Canopus has shown a steady decrease in water level (Fig.2). The ponds throughout the McMurdo Dry Valleys are undergoing both water level fluctuations, both increasing and decreasing, depending on their location and exposure to light. The water levels of the lakes and ponds in the McMurdo Dry Valleys are driven by the processes of ablation and evaporation (Fountain, et al., 2014; Hawes and Jungblut, 2016). These changes present impacts on microbial communities through either steep decreases or increases in planktonic production, changes in nutrient concentrations, and the loss of photosynthetic organisms found in the deeper layers of the water column (Hawes and Jungblut, 2016). The effect of sediment gain, and changes in ice-cover transparency may also change the community structure of the microbial mats (Hawes, et al., 2011).


Fig 2 – Water level change of Lake Canopus, McMurdo Dry Valleys, Victoria Land, Antarctica, as recorded by aerial surveys: (A) 1964; (B) 1970; (C) 1980; (D) 2004; (E) 2009; (F) 2016. Images show the continued decrease in the water level of Lake Canopus from 1964 – 2016 field seasons.



These limnetic systems act as natural laboratories, allowing us to take a unique look at how they are reacting to different shifts in water level changes. In our study, we aim to explore different microbial community structures at current water levels in order to create a baseline for future work. While the ecosystems in our study are not identical to each other we plan to set the frame work for understanding how these microbial communities are structured in a myriad of different water levels and ecosystems. This foundation will allow us to study the differences in the maintenance of community richness versus the structural richness already present in the mats. Our null hypothesis is that there will be no change in microbial mat community structure through the range of water levels and aquatic ecosystems.



2.1 Study Site


The McMurdo Dry Valleys host unique ecosystems which are sensitive to physical disturbances  by humans (Kong, et al., 2012). Because of this, the McMurdo Dry Valleys are protected under the Antarctic Treat as an Antarctic Special Managed Area (Kong, et al., 2012). An increase in economic activity, including tourism and scientific research, and climate change may have a heavy impact on the continent (Chown, et al., 2015). Attempts are being made to keep scientific sampling to an invasive minimum. Therefore, the samples used in this project have been provided from a collection of previous sample collection trips.




Fig 3 – (A) Map of the Antarctic continent, with the location of the McMurdo Dry Valleys, South Victoria Land highlighted by the red box; (B) The McMurdo Dry Valleys with the Wright Valley highlighted by the red box; (C) Sample collection sites in the Upper (3 sample ponds – L8, L16, and L26) and Central (Lake Canopus) Wright Valley.


Samples were collected from one lake and three ponds in the Wright Valley of the McMurdo Dry Valleys (Fig. 3 – A, B). Lake Canopus (77.54630, 161.53065) is located in the Central Wright Valley ~65 meters from the south shore of Lake Vanda. Aerial surveys have shown Lake Canopus to have continued water loss since 1964 (Fig. 2). Samples from Lake Canopus were collected along the paleo-shore lines of previous water edge, as well as at the highest point of the lake and the outer ridge of the lake where the water level was no longer present at the time of collection. The samples from the ponds were collected, between January 6th-8th 2017, from 3 varying depth levels of the ponds. The ponds, L8 (77.54620, 160.73813) (Fig 4), L16 (77.54139, 160.7555) (Fig 5), and L26 (77.55089, 160.72333) (Fig 6) are located in the Upper Wright Valley (Fig 3 – C) Samples were stored in sterile 50 mL falcon tubes. Samples from Lake Canopus were naturally dried, and samples from pond L8, L16, and L26 were freeze dried. Maps of the collection sites and sample locations were created using the Quantarctica package (Matsuoka, et al., Quantarctica, Norwegian Polar Institute), in QGIS (QGIS Development Team,, and Google Maps (Google Maps, 2018)

Fig 4 – Pond L8, Frozen, 2017


Fig 5 – (A) Pond L16, Partially Frozen, 2004; (B) Pond L16, Partially Frozen, 2016


Fig 6 – (A) Pond L26, Liquid, 2004; (B) Pond L26, 2017

2.2 DNA Extraction


We analyzed the microbial community structure of the sampled ponds and Lake Canopus through high-throughput sequencing of 16S (Prokaryote) rDNA. DNA was extracted using a PowerBiofilm DNA Isolation Kit (Mo BIO Laboratories, Carlsbad, CA) in accordance with the manufacturer’s instructions. Samples collected from Lake Canopus were collected as singular samples so each sample was mixed and DNA was extracted three times from the samples for duplicate purposes. Since the samples were freeze-dried, BF1 and BF2 were upped to the appropriate proportions when necessary, in accordance with the given protocol (a table is provided in the supplementary information for reference). Samples (n=4) with low or negative DNA concentrations were run with the provided troubleshooting protocol for DNA concentration. A NanoDrop 8000 Spectrophotometer (ThermoFisher Scientific, Waltham, MA) was used to quantify DNA yield (Table 1).


Table 1 – Final DNA concentration of all samples, 4 samples: Cano.4.A; Cano.4.B; Cano.4.C; Cano.Highest, were removed from the sample set after DNA amplification yielded no results even after the DNA concentration troubleshooting protocol was followed



2.3 DNA Amplification


Equipment was first sterilized under a laminar hood with UV lights for one hour to prevent contamination when creating the reaction mix. PCRs were run using a reaction volume totaling 19µl. A combination of 4µl of 5x GoTaqBuffer reaction buffer, 0.8µl of 20 mg/ml bovine serum albumin (BSA), 2µl of 25 mM MgCl2, 0.16µl of dDNTP, 0.2µl of Gotaq G2 DNA polymerase, 1µl of 10 µM forward primer, and 9.84µl of sterile PCR-grade water was used for each sample. A total of 18µl of the reaction mix and 1µl of the barcoded reverse primer was used per reaction tube. The PCR amplification of 16S samples was run using primers with MiSwq sequencing adapters, 12-nucleotide Golay barcoded reverse primers, and a forward prime. Each sample received a unique Golay barcode in order to allow the demultiplexing of the pooled samples after sequencing. A 515 forward primer (5’ – GTGCCAGCMGCCGCCGCGGTAA -3’) and its reverse complement 806R (3’- CAAGCAGAAGATACGAGAT -5’) were used (Caporaso, et al., 2011).

A triplicate of three different concentrations of DNA (0.5,1, and 1.5µl) and a negative control were run for each of the samples. The three concentrations of DNA were run in order to avoid amplification bias. The DNA was then amplified in a thermocycler (ThermoFisher Scientific, Waltham, MA) following the protocol of: 94°C for three min, a set of 30 cycles of 94°C for 45s, 50°C for 60s, and 72°C for 90s, followed by a final extension at 72°C for 10 min for 16S DNA. 1% Agarose gels were used to visually confirm amplification.

DNA was purified using an AxyPrep Mag PCR clean-up kit, according to the manufacturers’ instructions (Axygen Biosciences, Union City, CA). A Qubit 2.0 Fluorometer (ThermoFisher Scientific, Waltham MA) was used to measure the concentrations, in duplicate, of the purified PCR samples.



2.4 Sequencing

After purification, samples were pooled for sequencing. 300ng per sample of purified 16S DNA was pooled into a sterilized tube. Results from the Qubit were used to calculate the amounts of amplicon to be pooled from each sample to ensure equal concentrations. Once purified and pooled, the PCR products were sent to the sequencing facility at the Natural History Museum for sequencing using the Illumina MiSeq platform (Illumina, San Diego, CA) (Caporaso, et al. 2012).


2.5 Analysis


QIIME (Quantitative Insights into Microbial Ecology, v.1.8) (Caporaso, et al. 2010) was used to analyze the Illumina data. The Golay barcodes allowed the sequences to be demultiplexed on the Illumina MiSeq platform. Through Qiime full length amplicon sequences were created by merging the paired-end reads. Qiime was also used to both identify and remove chimeric sequences through the reference of the 16S rRNA Greengenes 13_8 database (Desantis, et al 2006; Edgar, et al 2011).  This same database was used identify OTUs (Operational Taxonomic Units). OTUs were then picked using and open reference method, and chloroplasts and mitochondrial taxonomic units were removed. A detailed reference of the commands used is provided in the supplementary information.


I. Alpha diversity analysis


QIIME was used to compute all alpha diversities indices and to create rarefied OTU tables. A combination of alpha diversity and abundance indices were run in order to avoid the biases of each of the given indices. Due to the small sample size, samples were not rarefied so as to maintain data points from all locations. Initially a comparison of Shannon’s and Simpson’s indices were run. These indices quantify abundance and evenness throughout the OTUs. Chao1 and ACE indices were used to quantify the taxa richness of OTUs from individual samples. Charts of each of the indices were run in PAST (v.3) (Hammer, et al., 2001) to visualize comparisons. A good’s coverage test was run to identify the number of species present in each individual sample. Sampling coverage was plotted using rarefaction curves for both Chao1 and number of observed species. Species abundance, in percentages, were plotted in Excel using the output from the OTU taxa summary plots previously achieved using QIIME.

II. Beta diversity Analysis

All beta analysis was run using PAST (v.3), using the rarefied OTU tables produced using QIIME, unless otherwise stated. Principal Component Analysis (PCA) was used to explore the variation between all locations at the phyla level as well as on two sites with the largest sampling numbers (L16 and Canopus). Due to their dominance, PCAs were also run at the order and genus levels of cyanobacteria and proteobacteria differences, between the sites. ANOSIM (Non-parameteric analysis of similarity) was run with Bonferroni corrections to explore the dissimilarities, using Bray-Curtis, between locations at the phyla level. This test was also run at the order and genus levels with a focus on cyanobacteria and proteobacteria. A SIMPER non-parametric test was then used explore the focal dissimilarities of the sites at a phyla level as well as at the order and genus level, again with a focus on cyanobacteria and proteobacteria.



3.1 Alpha Diversity

From the 21 16S rDNA samples a total OUT count of 2470300 was achieved with a mean of 117633.333. The outlier, sample 164 had a total OUT count of 28, but was included to analyze all depth layers from the L26 pond. Sample 164 (L26.3) was a sample with low DNA concentration which may have affected these results. Good’s coverage values of species richness was used to confirm good sampling coverage, all samples, with the exception of 164 (L26.4) were confirmed. Multiple alpha diversity tests were run to explore species diversity and richness throughout the samples (Table 2).

Sample Observed  Species Simpson Shannon Chao1 ACE Goods Coverage
Cano.1.A 2389 0.981453939 7.555192433 3072.280285 3266.471328 0.993508271
Cano.1.B 2129 0.96633374 7.057458712 2920.094276 2934.247438 0.995008622
Cano.1.C 2603 0.970310184 6.884409257 3944.607595 4084.067565 0.993009556
Cano.2.A 1524 0.989233824 7.928836989 2016.134529 2057.474964 0.99020038
Cano.2.B 2692 0.971292736 6.930577467 4050.12933 4266.065253 0.991050735
Cano.2.C 2511 0.957769389 6.613561614 3622.44186 3763.644181 0.995081124
Cano.3.A 2365 0.966031533 7.050761301 3556.324786 3643.391472 0.992990439
Cano.3.B 2492 0.985638242 7.662856329 3848.096 4076.613287 0.990853963
Cano.3.C 3121 0.988221773 7.914712678 4389.535294 4645.041896 0.993525078
Cano.Mat 1951 0.970072442 6.699708379 2577.083333 2798.128307 0.994300116
L8.A 953 0.816440849 4.378121598 1369.903226 1438.679955 0.996617749
L8.B 1103 0.961342758 6.139013722 1567.403974 1592.76071 0.995799543
L8.D 1420 0.882145304 5.282103777 2050.290837 2224.295484 0.995759712
L16.A 2742 0.964811901 6.649779447 4203.904255 4199.02906 0.995272027
L16.1 1258 0.960369171 5.967195856 1994.727749 2111.709576 0.994896096
L16.2 728 0.946454481 5.684179668 1183.640625 1051.600583 0.996555307
L16.3 1240 0.923450807 5.190401449 1755.204444 1913.123648 0.996233021
L16.Edge 1131 0.92701978 4.862610154 1738.6875 1788.665605 0.996336388
L26.1 1939 0.982317514 7.399242476 2384.031746 2481.798598 0.9954447
L26.3 23 0.951530612 4.450212065 48.5 64.4 0.357142857
L26.5 919 0.93073671 5.305064402 1315.290323 1391.856697 0.995559491

Table 2 – Total results for Alpha indices calculated using QIIME. Good’s coverage is a score based index for sampling coverage with a score of 0 meaning there is no coverage and a score of 1 represent complete coverage.


A comparison of Simpson’s and Shannon’s indices of species diversity was run as confirmation of alpha diversity results (Fig. 7). A similar comparison was run to confirm ACE and Chao1 richness indices (Fig. 8). Both comparisons show similar trends in overall abundance and richness confirming the output results.



Figure 7 – Comparison charts of (A) Shannon and (B) Simpson’s indices results. The graphs show similar trends in overall diversity. For Shannon indices, higher scores represent higher amounts of diversity within the sample. Simpson’s indices are score based, with 0 represent no diversity and 1 representing infinite diversity.


Figure 8 – Comparison charts of (A) Chao1 and (B) ACE indices results. The graphs show similar trends in overall species richness. The Chao index outputs assess both single and doubleton OTUs, while the ACE index asses all species containing less than 10 individuals.


The mean of each of the indices was calculated to identify the locations with the highest levels of richness and diversity. Lake Canopus showed the highest levels in both diversity and richness (Fig. 9, 10). Pond L8 showed the lowest levels of species diversity, while Pond L26 showed the lowest levels of species richness.

Figure 9 – Mean species diversity scores for each of the sampling sites. Shannon mean indices: 7.22980752 (Canopus); 5.26641303 (L8); 5.67083331 (L16); and (5.71817298). Simpsons mean indices: 0.97463578 (Canopus); 0.88664297 (L8); 0.94442123 (L16); and 0.95486161 (L26).


Figure 10 – Mean species richness scores for each of the sampling sites. Chao1 mean indices: 3399.67273 (Canopus); 1662.53268 (L8); 2175.23291 (L16); and 1249.27402 (L26). ACE mean indices: 3553.51457 (Canopus); 1751.91205 (L8); 2212.82569 (L16); 1312.6851 (L26).

Rarefaction curves were created to compare individual samples to the richness measure Chao1 and the total number of observed species (Fig. 11, 12). Both graphs show there to be good estimates of species richness within individual samples.


Figure 11 – Rarefaction curve for individual samples in comparison to the Chao1 index. Each colored line represents an individual sample, and the curve represents richness.  All curves reaching horizontal asymptote are good estimates.



Figure 12 – Rarefaction curve for individual samples in comparison to the number of observed species. Each colored line represents an individual sample, and the curve represents richness.  All curves reaching horizontal asymptote are good estimates.

Each sample was examined at the phyla level through abundance charts (Fig. 13). Proteobacteria and cyanobacteria are the dominant members of the microbial mat communities. Cyanobacteria makes up 20.8% of the community and proteobacteria is the most dominant of the phyla, making up 26.8% of the community. Other significant contributors are bacteroidetes (13.4%), and actinobacteria (8.7%).



Figure 13 – Abundance chart showing the occurrence percentage of each identified phyla in individual samples. Each color represents an identified phyla, while each bar is representative of a sample.



Due to their high occurrence levels, proteobacteria and cyanobacteria were explored through abundance charts on an individual basis at both the order (Fig. 14) and genus (Fig. 15) levels. Proteobacteria showed higher levels of diversity with 39 orders identified, and 107 genera identified. Cyanobacteria results identified 7 orders and 10 genera.



Figure 14 – Cyanobacteria abundance at the order level. Each bar represents a sampled site and each color represents an order.

At the order level, there is clear dominance of Synechococcophycideae pseudanabaenale(17.3%). Oscillatoriophycideae oscillatoriales (2.4%) has the highest presence in two locations from pond L16, as well as one sample from pond L16. There is surprisingly low presence of cyanobacteria in three samples from Lake Canopus: Cano.1.A, Cano.1.B, and Cano.2.A.



Figure 15 – Proteobacteria abundance at the order level. Each bar represents a sampled site and each color represents an order.


Proteobacteria have a higher diversity in the orders present and an increase in the admixture of dominance. Orders noted as dominant are Sphingomonadales(2.8%)Burkholderiales(6.5%), Myxococcales(1.9%), Rhodobacterales (4.1%), and Caulobacterales(1.6%).

Finally, proteobacteria (Fig. 16) and cyanobacteria (Fig. 15) were examined at the genus level for abundance. This becomes difficult as not all genera, particularly within cyanobacteria, have been classified.



Figure 15 – Cyanobacteria abundance at the genus level. Each bar represents a sampled site and each color represents an order.

There is once again clear dominance in the cyanobacteria with both dominant genera being of the family Psuedanabaenacea in the genera leptolyngbya (10.2%)and other (6.1%). The other most prominent genus is phormidiaceae (0.3%).




Figure 16 – Proteobacteria abundance at the genus level. Each bar represents a sampled site and each color represents an order.

Proteobacteria show high levels of admixture in dominance, and more variance between community structures from each sample and site. Dominating genera are: Marinicellaceae other (0.5%), Hahellaceae hahella (0.7%), Comamonadaceae other (3.6%), Rhodobacteraceae anaerospora (1.9%), Caulobacteraceae mycoplana (1.6%)and Sphingomonadaceae kaistobacter (1.2%).

3.2 Beta Analysis

An initial PCA was run on all samples to explore phyla level clustering and to identify any unique groupings for further analysis (Fig. 17, Table 3). PCAs were also run on the two sites with the largest collection of samples, Lake Canopus (Fig. 18, Table 4) and pond L16 (Fig. 19, Table 5)



PC Eigenvalue % Variance
1 251.094 56.921
2 88.8221 20.135
3 32.864 7.45
4 27.4667 6.2265
5 18.2968 4.1477


Figure 17; Table 3 – Principal Component Analysis clustering chart showing all locations based on phyla. Eigenvalues displays clustering of groups by sample location.




PC Eigenvalue % Variance
1 176.226 55.057
2 83.9298 26.221
3 45.8715 14.33


Figure 18; Table 4 – Principal Component Analysis clustering chart showing all Lake Canopus samples based on phyla. Eigenvalues indicate significant clustering of groups by sample location.


PC Eigenvalue % Variance
1 141.381 50.824
2 77.7922 27.965
3 53.6316 19.28
4 5.3697 1.9303


Figure 19; Table 5 – Principal Component Analysis clustering chart showing all L16 samples based on phyla. Eigenvalues indicate significant clustering of groups by sample location.

All PCAs at the phyla level show significant clustering due to variance in location. The comparison of all locations shows Lake Canopus to be clearly distinct in community structure. Similarly, there are distinctions in depth and location within the sample sites of Lake Canopus and pond L16. Pond L16 in particular shows clear separation from the core mat samples (L16.1, L16.2, and L16.3) from those collected around the ponds outer edges (L16.A and L16.Edge).


PC Eigenvalue % Variance
1 25.1499 42.139
2 11.8921 19.926
3 6.95753 11.658


Figure 20; Table 6 – Principal Component Analysis clustering chart showing all sample sites based on proteobacteria genus. Eigenvalues indicate clustering of groups by sample location.


PC Eigenvalue % Variance
1 159.508 62.45
2 74.057 28.994
3 14.6991 5.7549


Figure 21; Table 7 – Principal Component Analysis clustering chart showing all site samples based on cyanbacteria genus. Eigenvalues indicate clustering of groups by sample location.


PCA analysis of proteobacteria (Fig. 20, Table 6) and cyanobacteria (Fig. 21, Table 7) at the genus level show that there is still slight clustering of groups. However, this clustering becomes much less distinct, especially in the case of cyanobacteria.

ANOSIMs, from Bray-Curtis dissimilarities, were run with Bonferroni corrections to compare dissimilarities between all locations (Table 8, 9, 10).

A 1 3 4 2
1 0.0349 0.0514 0.0294
3 0.0349 0.6978 0.4269
4 0.0514 0.6978 0.9025
2 0.0294 0.4269 0.9025
B 1 3 4 2
1 0.2094 0.3084 0.1764
3 0.2094 1 1
4 0.3084 1 1
2 0.1764 1 1


Table 8 – ANOSIM results from Bray-Curtis dissimilarities at the phyla level: (A) p-values, uncorrected; (B) Bonferroni-corrected p-values. Site locations: (1) Lake Canopus; (2) L8; (3) L16, (4) L26. Values of significance are highlighted in green.


A 1 3 2 4
1 0.0051 0.014 0.0228
3 0.0051 0.6279 0.7222
2 0.014 0.6279 0.2977
4 0.0228 0.7222 0.2977


B 1 3 2 4
1 0.0306 0.084 0.1368
3 0.0306 1 1
2 0.084 1 1
4 0.1368 1 1



Table 9 – ANOSIM results from Bray-Curtis dissimilarities at the order level: (A) p-values, uncorrected; (B) Bonferroni-corrected p-values. Site locations: (1) Lake Canopus; (2) L8; (3) L16, (4) L26. Values of significance are highlighted in green.


A 1 3 4 2
1 0.0024 0.0058 0.0103
3 0.0024 0.6835 0.4303
4 0.0058 0.6835 0.5024
2 0.0103 0.4303 0.5024


B 1 3 4 2
1 0.0144 0.0348 0.0618
3 0.0144 1 1
4 0.0348 1 1
2 0.0618 1 1



Table 10 – ANOSIM results from Bray-Curtis dissimilarities at the genus level: (A) p-values, uncorrected; (B) Bonferroni-corrected p-values. Site locations: (1) Lake Canopus; (2) L8; (3) L16, (4) L26. Values of significance are highlighted in green.


After the implementation of Bonferroni corrections, dissimilarities between locations becomes more apparent at the genus level. At the phyla level, after correction were made, there are no significant dissimilarities. ANOSIM at the order level shows dissimilarities between L16 and Lake Canopus. It is at the genus level that we see the most significant dissimilarities between both L16 and L8 in comparison with Lake Canopus.

SIMPER analysis of Bray-Curtis dissimilarities were run for all locations at the phyla, order and genus levels. For analysis purposes, we have focused on the top five contributors to dissimilarities between locations at the phyla (Table 11) and order (Table 12) levels as the identification of taxa at the genera is limited.

Taxon Av. dissim Contrib. % Cumulative %
GN02 20 43.45 43.45
Cyanobacteria 5.245 11.4 54.85
Chloroflexi 4.092 8.891 63.74
Firmicutes 3.112 6.761 70.5
Bateroidetes 3.041 6.606 77.11


Taxon Av. dissim Contrib. % Cumulative %
Cyanobacteria 9.913 26.16 26.16
Firmicutes 4.597 12.13 38.3
Proteobacteria 3.998 10.55 48.85
Chloroflexi 3.931 10.37 59.22
Actinobacteria 3.479 9.183 68.4


  Taxon Av. dissim Contrib. % Cumulative %
Cyanobacteria 14.45 37.55 37.55
Firmicutes 4.581 11.91 49.46
Chloroflexi 4.288 11.15 60.6
Proteobacteria 4.272 11.1 71.7
Bateroidetes 2.883 7.491 79.2



  Taxon Av. dissim Contrib. % Cumulative %
GN02 20 46.41 46.41
Cyanobacteria 7.187 16.68 63.08
Bateroidetes 2.851 6.616 69.7
Planctomycetes 2.72 6.311 76.01
Chloroflexi 2.242 5.202 81.21


  Taxon Av. dissim Contrib. % Cumulative %
GN02 20.01 46.28 46.28
Cyanobacteria 3.192 7.382 53.66
Bateroidetes 3.156 7.299 60.96
Planctomycetes 2.931 6.777 67.74
Chloroflexi 2.7 6.244 73.98


  Taxon Av. dissim Contrib. % Cumulative %
Cyanobacteria 7.001 25.68 25.68
Chloroflexi 3.009 11.03 36.71
Bateroidetes 3.003 11.01 47.73
Actinobacteria 2.758 10.12 57.84
Proteobacteria 2.703 9.913 67.76



Table 11 – SIMPER analysis of top phyla level contributors to Bray-Curtis dissimilarities. (A) Lake Canopus – L16; (B) Lake Canopus – L26; (C) Lake Canopus – L8; (D) L16 – L8; (E) L16 – L26; (F) L8 – L26.

The top contributors to dissimilarities between all locations show repetitions of the same phyla, with the greatest contributors being cyanobacteria and GN02.

  Taxon Av. dissim Contrib. % Cumulative %
Pseudanabaenales 13.95 22.05 22.05
Ellin6067 10.04 15.88 37.93
Burkholderiales 8.122 12.84 50.77
Oscillatoriales 6.046 9.56 60.33
Rhodobacterales 5.856 9.26 69.59
  Taxon Av. dissim Contrib. % Cumulative %
CV90 31.7 42.31 42.31
Pseudanabaenales 17.25 23.02 65.33
Ellin6067 8.358 11.15 76.48
Burkholderiales 3.877 5.174 81.66
Caulobacterales 2.808 3.748 85.4
  Taxon Av. dissim Contrib. % Cumulative %
Pseudanabaenales 15.36 23.74 23.74
Ellin6067 10.04 15.52 39.26
Burkholderiales 8.293 12.82 52.08
Myxococcales 4.277 6.61 58.69
Rhodobacterales 3.487 5.389 64.08
  Taxon Av. dissim Contrib. % Cumulative %
CV90 33.36 52.69 52.69
Pseudanabaenales 10.92 17.25 69.94
Oscillatoriales 3.582 5.657 75.59
Rhodobacterales 2.732 4.314 79.91
Burkholderiales 2.264 3.575 83.48
  Taxon Av. dissim Contrib. % Cumulative %
CV90 33.37 55.54 55.54
Pseudanabaenales 8.227 13.69 69.24
Caulobacterales 2.738 4.557 73.79
Myxococcales 2.044 3.403 77.2
Oscillatoriales 1.947 3.241 80.44
  Taxon Av. dissim Contrib. % Cumulative %
Pseudanabaenales 6.716 16.32 16.32
Oscillatoriales 5.818 14.14 30.47
Rhodobacterales 5.216 12.68 43.15
Myxococcales 3.415 8.3 51.45
Sphingomonadales 3.029 7.361 58.81


Table 12 – SIMPER analysis of top order level contributors to Bray-Curtis dissimilarities. (A) Lake Canopus – L16; (B) Lake Canopus – L26; (C) Lake Canopus – L8; (D) L16 – L8; (E) L16 – L26; (F) L8 – L26.


Repetitions in the top contributors of dissimilarity between locations at the order level are also apparent. The top contributors are: psuedanabaenales, oscillatoriales, and rhodobacterales.



Microbial mats from four meltwater ponds and one lake from the McMurdo Dry Valleys were sequenced to explore the structure of mat communities and their responses to continued water level changes. Analysis was conducted at the phyla, order and genus level for all location. Due to the lack of identified species we were unable to explore species diversity of cyanobacteria between the sample sites.

Our findings show there to be dominance by proteobacteria and cyanobacteria in the community structure at all sampled locations. The abundance of identified cyanobacteria and proteobacteria orders is consistent with previous research by Zhang et al. (2015), Quesdada et a.l (2012), and Hawes et al. (2011). Zhang et al (2015) received similar community structure results in the analysis of several lakes from the McMurdo Dry Valleys. They suggest the lack of diversity within samples taken from under the ice may be due to the lack of Nostocales. All samples from our sites collected from below the ice sheets have little to no presences of Nostocales, supporting this hypothesis. This theory is further supported by the outlying samples collected form the end of L16 and Lake Canopus which showed markedly different community structure suggesting that ice cover and exposure to light are major factors in determining microbial mat community composition. Apart from sample L26.3 all samples showed similar community structures at the phyla level. It is at genus level where we see statistically significant variances between pond and lake community samples. In their study Jungblut et al. (2005) found that ponds from the McMurdo Dry Valleys within close proximity to each other showed differences in their microbial mat communities; our results support their work.

Quesada et al (2008) argues in their paper that the benthic communities from Antarctic lakes would have variable physiological responses to immediate environmental pressures creating short term variability within the layers but with the community structure, remaining relatively constant in yearly cycles. The high abundance of many of the same taxa within the samples suggests there is a metapopulation that remains relatively constant throughout the microbial mats in the Wright Valley of the McMurdo Dry Valleys. Within these metapopulations there is significant diversity in the genera present. Junglbut et al (2017b) suggests metapopulations developed due to the highly stressful environment, and the taxa that thrive there are relatively constant due to their adaptations to survive these conditions. Having a constant metapopulation with diversity within the genera would allow the mats to be both resistant and resilient to on-going ecosystem pressure at the phyla level but to continue to adapt to the changing pressures through shifts in community structure at the genera level.

Analysis of both L16 and Lake Canopus show there to be clear separations within the sample sites. The clustering shown on both PCAs displays similar results with samples from the centers of the pond clustering together and samples from the outer edges of site acting as outliers. Unfortunately, DNA extraction the outer most quadrat from Lake Canopus was unsuccessful. A previous study by Hawes and Jungblut (2016) looked at the impacts of irradiance on the photosynthetic rates of mat communities. They found that high levels of exposure to irradiance led to lower levels of community growth within the mat (Hawes, et al., 2011). This may be why we see outliers from the outer edges of sample sites as well as the upper layers clustering separately from the middle and lower levels. Hawes and Jungblut (2016) also showed that differing levels of ice transparency accounted for levels of thickness in the laminal mat layers. This may most heavily impact the samples on the outer edges of the sites as they have the least ice coverage and are therefore exposed to more intensive pressures including irradiance. However, it is also possible that the outlying samples were under little to no water coverage and have not been photosynthetically active for an extended period of time, causing differences to appear in the community structure as the less resilient taxa die off.

Differences between layers of the microbial mats were present, but more work is needed in order to better understand the drivers of these differences. We suggest that future sampling efforts take depth level samples at multiple locations throughout the individual sites. Both the separation of the edge communities in L16 and Lake Canopus sites and the clustering of the smaller, layered groups shows there are differences in the levels of community structure but more samples and environmental recordings are needed. Future collection expeditions should make efforts to collect samples from the same water levels depths at all sites for equal comparisons.

We call for continued research, when possible, of historic and modern samples of microbial mat communities, such as the recent work by Jungblut et al (2017b). Studies of this type will allow us to track the introduction of exotic taxa, as well as creating a time-frame for the dynamic structure of community diversity and its interaction with environmental and climactic changes. As keystone taxa, the effects of climate change on cyanobacterial communities will have major impacts on ecosystem functions, food webs, and carbon cycling (Jungblut, et al., 2017B)

In the future, we suggest the sample sites from this study be monitored for environmental and geographic data to create a cohesive picture of how microbial mats are evolving and adapting to water level changes and current climatic shifts. Since the structure of microbial mat communities is determined by their environment, records should be kept on the sites salinity, pH, temperature, and water level when future sampling is conducted. Research by Verleyen et al., (2010) shows salinity to be one of the most important factors in determining microbial mat community structure, contradicting previous microscopy studies which identified pH and lake depth as the most important factor. We therefore encourage future research to continue to explore the role of the environment in structuring communities.

Several climate models have predicted continued warming in the Polar Regions, and as the ozone hole continues to fill the warming of the Antarctic continent is expected to accelerate (Jungblut, and Hawes, 2017; Fountain et al., 2014). To better understand the long-term implications of climatic-driven changes we must continue to study the time-scale of population drift, community structure and the diversity of microbial communities and the forces that shape them (Jungblut et al, 2018).

Cyanobacteria are producers of cyanotoxins, a diverse range of natural toxins known to include several mechanisms of toxicity including tumor promotion, neurotoxicity, and genotoxicity (Jungblut et al., 2018). A study by Junblut et al (2018) showed that 100-year-old samples still contained viable cyanotoxins, showing they are resistant to intensive pressures like their producers. Little is known about the mechanisms of cyanotoxin production and its role in freshwater ecology. It has been shown that cyanotoxin production is mitigated by environmental conditions (Jungblut, et al., 2018). Work by Kleinteich et al (212) suggests that the continued increase in water temperatures throughout Antarctica could favor heightened production of cyanotoxins, making the study of these toxins and their producers of utmost importance for global public health.


Due to the time frame of this project, we were unable to analyze the sequenced 18S (Eukaryote) data. Therefore, it is suggested that this analysis is continued. Culturing of all sample sites was also conducted. While the culturing of several of the samples were unsuccessful due to their being freeze-dried after collection, two sample sites from Lake Canopus showed successful regrowth. The samples from Lake Canopus were naturally dried after their collection, and this may be indicative of strong resilience to water level changes. We strongly encourage future culturing work to be carried out on naturally dried samples to better explore this resilience and how it may impact the community structure of the microbial mats. Our results show that cyanobacteria inhabiting microbial mats are resilient to water level changes. Culturing of the naturally dried samples showed successful re-growth and phyla community structure was similar between different levels of submergence, showing the differing water levels do not heavily impact the community structure at this level.



Having evolved under the harsh conditions of the Precambrian, modern cyanobacteria are representatives of the microorganisms’ ability to adapt and survive in the most extreme environments in the world (Jungblut and Hawes, 2017). Cyanobacteria are keystone taxa in perennially cold environments and play a major role in ecological processes (Chrismas, et al., 2015; Jungblut and Hawes, 2017). With the modern pressures of climate change already effecting the polar regions it is more pressing now than ever to study these incredible microorganisms to track their ability to adapt to these changes and how these changes affect the neighboring ecosystems. Our study shows there are distinct community structural differences between sampled locations, but little variance within them. Due to outlier samples from the edge of two sample sites we also theorize that light exposure and ice cover play major roles in shaping the community structure of microbial mats. More research in needed on microbial communities throughout the Antarctic so we may better predict the effects of climate change by creating a baseline of current microbial diversity and community structures (Verleyen, et al., 2010).


Special thanks to Ian Hawes, Stephania Tsola, and Meagan Bell for their contributions towards this project.



















Allison, S., Martiny, J. (2008) Resistance, resilience, and redundancy in microbial communities. Proceedings of the National Academy of Sciences. 105; 11512-11519. DOI: 10.1073/pnas.0801925105

Archer, S., McDonald, I., et al. (2014) Characterisation of bacterioplankton communities in the meltwater ponds of Bratina Island, Victoria Land, Antarctica. Microbiology Ecology. 89; 451-464. DOI: 10.1111/1574-6941.12358

Caporaso, J., Kucaynski, J., et al (2010) QIIME allows analysis of hight-throughput community sequencing data. Nature Methods. 7; 335 DOI: 10.1038/nmeth.f.303

Caporaso, J., Lauber, C. et al. (2011) Global patterns of 16S rRNA diversity at a depth of millions of sequences per sample. Proceedings of the National Academy of Sciences. 108; 4516-4522 DOI 10.1073/pnas.1000080107

Caporaso, J., Lauber, C., et al. (2012). Ultra-high-throughput microbial community analysis on the Illumina HiSeq and MiSeq platforms. The ISME Journal. 6; 1621-1624. DOI: 10.1038/ismej.2012.8

Chown, S., Clarke, A., et al. (2015) The changing form of Antarctic biodiversity. Nature. 522; 431-438. DOI: 10.1038/nature14505

Chrismas, N., Eanesia, A., et al (2015) Multiple adaptations to polar and alpine evironments within cyanobacteria: a phylogenomic and Beysian approach. Frontiers in Microbiology. 6; 1070-1070 DOI: 10.3389/fmicb.2015.01070

Chrismas, N., Barker, G., et al (2016) Genomic mechanisms for cold tolerance and production of exopolysaccharides in the Arctic cyanobacterium Phormidesmis priestleyi BC 1401. BMC genomics. 17; 533 DOI: 10.1187/s12864-01602846-4

Desantis, T., Hugeholtz. P., et al (2006) Greengenes, a Chimera-Check 16s rRNA Gene Database and Workbench Compatible with ARB. Applied Environmental Microbiology. 72; 5069-5072 DOI 10.1128/AEM.03006-05

Doran, P., Priscu, J., et al. (2002) Antarctic climate cooling and terrestrial ecosystem response. Nature. 415; 517-520 DOI: 10.1038/nature710

Edwards, A., Pachebat, J. (2013) A metagenomics snapshot of taxonomic and functional diversity in an alpine glacier cryoconite ecosystem. Environmental Research Letters. 8 DOI: 10.1088/1748-9326/8/035003

Edgar, R., Haas, B., et al (2011) UCHIME improves sensitivity and speed of chimera detection. Bioinformatics. 27; 2194-2200 DOI: 10.1093/bioinformatics/btr381

Fountain, A., Levy, J., et al (2014) The McMurdo Dry Valleys: A landscape on the threshold of change. Geomorphology 225; 25-35 DOI: 10.1016/j.geomorph.2014.03.044

Google Maps (2018) Google Maps. Online. Available at: Last Accessed: August 2, 2018.

Hammer, O., Harper, D., et al (2001) PAST: Paleontological statistics software package for education and data analysis. Paleontologia Electronica. 4; 9

Hawes, I., Schwarz, A-M. (2001a) Absorption and utilization of irradiance by cyanobacterial mats in two ice-covered Antarctic lakes with contrasting light climates. Journal of Phycology. 37; 5-15 DOI: 10.1046/j.1529-8817.1999.014012005.x

Hawes, I., Moorhead, D., et al (2001b) Benthic primary production in two perennial ice-covered Antarctic lakes: patterns of biomass accumulation with a model of community metabolism. Antarctic Science. 13; 18-27 DOI: 10.1017/S0954102001000049

Hawes, I., Sumner, D., et al (2011) Legacies of recent environmental change in the benthic communities of Lake Joyce, a perennially ice-covered Antarctic lake. Geobiology. 9(5); 394-410 DOI 10.1111/j.1472-4669.2011.00289.x

Hawes, I., Jungblut, A. (2016) Growth dynamics of a laminated microbial mat in response to variable irradiance in an Antarctic lake. Freshwater Biology. 61; 396-410 DOI: 10.1111/fwb.12715

Jungblut, A., Hawes, I., et al (2005) Diversity within cyanobacterial mat communities in variable salinity meltwater ponds of McMurdo Ice Shelf, Antarctica. Environmental Microbiology. 7(4); 519-529 DOI: 10.1111/j.1462-2920.2004.007117.x

Jungblut, A., Hawes, I., et al. (2015) Microbial mat communities along an oxygen gradients in a perennially ice-covered Antarctic lake. Applied and Environmental Microbiology. 82(2); 620-630. DOI: 10.1128/AEM.02699-15

Jungblut, A., Hawes, I. (2017) Using Captain Scott’s Discovery specimens to unlock the past: has Antarctic cyanobacterial diversity changed over the last 100 years? Proceedings Royal Society B 284: 20170833. DOI: 10.1098/rspb.2017.0833

Jungblut, A., Wilbraham, A., et al. (2018) Microcystins, BMAA and BMAA isomers in 100-year-old Antarctic cyanobacterial mats collected during Captain R.F. Scott’s Discovery Expedition. European Journal of Phycology. DOI 10.1080/09670262.2018.1442587

Kleinteich, J., Hildebrand, F., et al. (2014) Diversity of toxin and non-toxin containing cyanobacterial mats of meltwater ponds on the Antarctic Peninsula: a pyrosequencing approach. Antarctic Science. 26(5); 521-532 DOI: 10.1017/S0954102014000145

Kong, W., Ream, D., et al. (2012) Diversity and expression of RubisCO genes in perennial covered Antarctic lake during the polar night transition. Applied and Environmental Microbiology. 78(12); 4358-4366. DOI: 10.1128/AEM.00029-12

Matsuoka, K., Skoglund, A., et al. (2018) Quantarctica. Norwegian Polar Institute. DOI: 10.21334/npolar.2018.85163961

Miteva, V. (2008). Bacteria in snow and glaciers. Psychrophiles from Biodiversity to Biotechnology. Springer, Berlin. Heidelberg. 31-50 DOI: 10.1007/978-3-540-74335-4_3

Morgan-Kiss, R., Priscu, J., et al. (2006) Adaptation and acclimation of photosynthetic microorganisms to permanently cold environments. Microbiology and Molecular Biology Reviews. 70; 222-252 DOI: 10.1128/MMBR.70.1.222-252.2006

Oren A (2004) A proposal for further integration of the cyanobacteria under the Bacteriological Code. International Journal of Systematic and Evolutionary Biology. 54; 1895-1902 DOI: 10.1099/ijs.0.03008

QGIS Development Team (2018) QGIS Geographic Information System Open Source Geospation Foundation Project.

Quesada. A., Fernanadez-Valiente, E., et al (2008) Benthic primary production in polar lakes and rivers. Polar Lakes and Rivers: Limnology of Arctic and Antarctic aquatic ecosystems. Oxford University Press Inc., New York. 179-196 DOI: 10.1093/acprof:oso/9780199213887.001.0001

Shade, A., Peter, H., et al. (2012) Fundamentals of microbial community resistance and resilience. Frontiers in Microbiology. 3; 1-19 DOI: 10.3389/fmicb.2012.00417

Stanier R., Sistrom W., et al. (1978) Proposal to place nomenclature of cyanobacteria (blue-green-algae) under rules of International Code of Nomenclature of Bacteria. International Journal of Systematic and Evolutionary Biology. 28 335-336 DOI: 10.1099/00207713-28-2-335

Tang, E., Tremblay, R., Vincent, W. (1997) Cyanobacterial dominance of polar freshwater ecosystems: are high-latitude mat-formers adapted to low temperature? Journal of Phycology 33(2); 171-181 DOI: 10.1111/j.0022-36-46.1997.00171.x

Tashyreva, D., Elster, J. (201) Annual cycles of two cyanobacterial mat communities in hydro-terrestrial habitats of the High Arctic. Microbial Ecology 71; 887-900 DOI: 10.1007/s00248-016-0732-x

Varin, T., Lovejoy. C., et al. (2012) Metagenomic analysis of stress genes in microbial mat communities from extreme Arctic and Antarctic environments. Applied Environmental Microbiology 78; 549-559 DOI: 10.1128/AEM.06354-11

Velázquez, D., López- Bueono, A., et al. (2016) Ecosystem function decays by fungal outbreaks in Antarctic microbial mats. Scientific Reports 6:22954. DOI: 10.1038/srep22954.

Verleyen, E., Sabbe, K., et al (2010). Structuring effects of climate-related environmental factors on Antarctic microbial mat communities. Aquatic Microbial Ecoloby. 59; 11-24 DOI: 10.3354/ame01378

Vincent, W. (2007) Cold tolerance in cyanobacteria and life in the cryosphere. Algae and cyanobacteria in extreme environments Cellular Origin, Life in Extreme Habitats and Astrobiology. 11. Springer, Heidelberg, pp 287-301 DOI 10.1007/978-1-4020-6112-7_15

Vincent, W., Quesada, A. (2012) Cyanobacteria in high latitude lakes, rivers and seas. Ecology of Cyanobacteria II: Their Diversity in Space and Time. Springer Science + Business Media. DOI: 10.1007/978-94-007-3855-3_13

Zhang, L., Jungblut, A., et al (2015) Cyanobacterial diversity in benthic mats of the McMurdo Dry Valley lakes, Antarctica. Polar Biology. 38(8); 1097-1110. DOI 10.1007/s003000-015-1669-0
























Supplementary Information


I. Buffer Concentrations

II. QIIME Commands



Buffer Concentrations used during DNA Extraction


Cano.1.A 350µl 100µl
Cano.1.B 350µl 100µl
Cano.1.C 350µl 100µl
Cano.2.A 350µl 100µl
Cano.2.B 350µl 100µl
Cano.2.C 350µl 100µl
Cano.3.A 350µl 100µl
Cano.3.B 350µl 100µl
Cano.3.C 350µl 100µl
Cano.4.A 700µl 200µl
Cano.4.B 700µl 200µl
Cano.4.C 700µl 200µl
L8.A 400µl 114µl
L8.B 400µl 114µl
L8.D 700µl 200µl
L16.3 400µl 114µl
L16.A 400µl 114µl
L16.1 400µl 114µl
L16.2 400µl 114µl
L16.3 700µl 200µl
L16.Edge 700µl 200µl
L.26.1 700µl 200µl
L26.5 400µl 114µl
L26.3 400µl 114µl
Cano.Highest 700µl 200µl
Cano.Mat 400µl 114µl





QIIME Commands
biom summarize-table




EssayHub’s Community of Professional Tutors & Editors
Tutoring Service, EssayHub
Professional Essay Writers for Hire
Essay Writing Service, EssayPro
Professional Custom
Professional Custom Essay Writing Services
In need of qualified essay help online or professional assistance with your research paper?
Browsing the web for a reliable custom writing service to give you a hand with college assignment?
Out of time and require quick and moreover effective support with your term paper or dissertation?
Did you find someone who can help?

Get quality dissertation help from certified experts, along with a free originality report attached.