Methodology:

Fellfield soils were sampled between November 2007 and February 2008 from 29 sites along a 1,650 km latitudinal transect between Signy Island in the South Orkney Islands (60 deg S) and south-eastern Alexander Island (72 deg S). The soils that were collected were devoid of vegetation and were hence representative of the soils that typically form in maritime Antarctica. The sampling sites were either accessed by small boat or helicopter deployed from a ship, by fixed wing aircraft fitted with skis, or by foot from research stations. At each of the sites, latitude, longitude and altitude were recorded and five samples of soil (0-50 mm depth, c. 50 cm3) were placed into DNA/RNAase-treated plastic tubes and bulked. The soils were frozen immediately at c. -80 deg C in a mixture of dry ice and ethanol and were kept at this temperature until they were processed.

Twenty edaphic factors were measured in each 4 mm-sieved soil, viz., moisture concentration, pH (H2O), electrical conductivity, the concentrations of total organic C and N, Ca, Cu, Fe, K, Mg, Mn, Ni, P, Zn, and those of water-extractable phosphate ion, sulphate ion, chloride ion, ammonium-N ion, nitrate/nitrite-N ion and dissolved organic carbon (DOC). Soil organic C:N ratio was also calculated. Soil pH and electrical conductivity were measured in 1:2.5 and 1:5 soil:water (vol:vol) slurries, respectively. Total nitrogen and organic carbon concentrations were determined using an elemental analyzer following drying at 105 deg C and treatment with HCl to remove inorganic carbon. Concentrations of Ca, Cu, Fe, K, Mg, Mn, Ni, P and Zn were determined using ICP-MS following reverse aqua regia digests. Water extractable phosphate ion, sulphate ion and chloride ion concentrations were determined using Dionex ion chromatography. Soil ammonium-N ion, nitrate/nitrite-N ions and dissolved organic carbon (DOC) concentrations were determined in 1M KCl (1:4 dry soil equivalent to solution ratio) followed by automated flow injection analysis. The moisture concentration of each soil sample was determined gravimetrically. All analyses were based on soil passed through a 4 mm sieve. Data were expressed as mg kg-1 dry soil, with the exception of moisture concentration (%), electrical conductivity (microsecond cm-1) and organic C:N ratio (dimensionless).

Data collection:

The following instruments were used in the analyses:

1. Total nitrogen and organic carbon concentrations: Exeter Analytical CE440 Elemental Analyzer (Coventry, UK).

2. Water extractable PO43-, SO42- and Cl- concentrations: Dionex ion chromatography (Sunnyvale, California, USA).

3. Ca, Cu, Fe, K, Mg, Mn, Ni, P and Zn concentrations: inductively coupled plasma mass spectrometry, following reverse aqua regia digests.

4. NH4+-N, NO3--N/NO2--N and dissolved organic carbon (DOC) concentrations: automated flow injection analysis (Skalar Analytical B.V., The Netherlands).

5. Electrical conductivity and pH value: Hanna Instruments pH meter (Leighton Buzzard, UK).

Data quality:

Instruments were calibrated according to laboratory practices, with standards being deployed where necessary. No data losses and no data cleaning.