Snow was collected from the surface to represent collection during melt water production for hot-water drilling. At each sampling location, a 1 m snow pit was excavated and the top 30 cm of snow was sampled using an ethanol sterilized shovel and Whirl-Pak bags (Nasco, WI, United States). Samples were transported to the field laboratory, where they were melted and passed through 0.2 μm Sterivex filters (Merck, Darmstadt, Germany), before freezing at −20°C for transport and processing in the United Kingdom.
Study Extent: Snow samples were collected between the South Orkney Islands and the Ellsworth Mountains between December 2012 and January 2014. On Signy Island, two sites were sampled; Gourlay Snowfield and Tuva Glacier. SkyBlu samples came from the vicinity of the BAS blue ice runway (a logistics hub for deep field operations). Camp samples were collected around the kitchen, generator and drilling areas.
Method step description:

1. SkyBlu, Pine Island Bay and Ellsworth snow samples were processed in two ways; PMA-treated and non-PMA-treated in order to differentiate the potentially viable microbial community from relic DNA. Each 0.2 μm filter was cut in half using sterile and DNAase/ethanol treated razors. For each sample, one-half was processed with PMA as per Nocker and Camper (2009) and Fittipaldi et al. (2012), using a 20 mM stock solution of PMA (Biotium, Hayward, CA, United States) in a 20% (v/v) aqueous solution of dimethyl sulfoxide (DMSO). Filters for PMA treatment were placed in a 6-well plate and a PMA solution at a final concentration of 100 μM was added. Cross-linking was initiated by 10 min incubation on ice, in the dark with occasional mixing, followed by 5 min of light exposure using a 650 W halogen lamp (FLASH 2000 L, DTS, Italy), at a 20 cm distance. The process was carried out in a laminar flow hood to avoid contamination of the samples. Non-PMA-treated samples were incubated in a 20% (v/v) aqueous solution of DMSO, and treated following the same protocol as PMA-treated samples. All samples were washed twice with sterile phosphate buffered saline (PBS) and all samples were then used for DNA extraction. DNA from snow samples was extracted using the PowerWater kit from MoBio (Qiagen, Carlsbad, CA, United States). Each sample was PCR amplified using the primers 341F and 785R covering the V3-V4 regions of the 16S rRNA gene (Klindworth et al., 2013), under the following conditions: initial denaturation at 95°C for 5 min then 25 cycles of 40 s denaturation at 95°C; primer annealing at 55°C for 2 min; and elongation at 72°C for 1 min then a final elongation at 72°C for 7 min (Hodson et al., 2017). DNA extraction kit controls were included alongside the snow derived DNA and sequenced. PCR amplicons were cleaned, normalized, quantified and supplemented with 5% PhiX before being loaded on Illumina MiSeq, as per the Illumina standard operating protocol (Kozich et al., 2013).
2. DNA from snow samples was extracted using the PowerWater kit from MoBio (Qiagen, Carlsbad, CA, United States). Each sample was PCR amplified using the primers 341F and 785R covering the V3-V4 regions of the 16S rRNA gene, under the following conditions: initial denaturation at 95°C for 5 min then 25 cycles of 40 s denaturation at 95°C; primer annealing at 55°C for 2 min; and elongation at 72°C for 1 min then a final elongation at 72°C for 7 min (Hodson et al., 2017). DNA extraction kit controls were included alongside the snow derived DNA and sequenced. PCR amplicons were cleaned, normalized, quantified and supplemented with 5% PhiX before being loaded on Illumina MiSeq, as per the Illumina standard operating protocol (Kozich et al., 2013).