The Geological Survey of Northern Ireland (GSNI) and the British Geological Survey (BGS) completed a regional geochemical survey of Northern Ireland's soils, sediments and waters between 2004 and 2006.

Soil sampling of all of Northern Ireland was completed under the Tellus survey between 2004 and 2006. Soil samples were collected on a systematic basis from rural areas in most of the region, excluding only the major urban centres of Belfast and Bangor. Soils were also collected at a higher sampling density from the urban areas of Belfast, Bangor, Carrickfergus, Carryduff, Castlereagh, Greenisland, Holywood, Lisburn, Newtownabbey and Londonderry, although these urban results are not reported here.

In rural areas, samples were collected from alternate 1 km Irish national- grid squares. Site selection within each square was random, subject to the avoidance wherever possible of roads, tracks, railways, human habitation and other disturbed ground. At each site two composite samples of five auger flights were collected, each composite sample comprising approximately 750 g of unsieved material. Samples were collected using a hand auger with a 20 by 5 cm flight from a standard depth interval of 5–20 cm for designated ‘A’ samples, referred to subsequently as ‘surface soils’, and at 35–50 cm for designated ‘S’ samples (nominally the B horizon), referred to subsequently as ‘deep soils’. Some 6,862 regional soil sites were sampled (see supplementary map - soil locations) and analysed, resulting in an average regional sampling density of 1 site per 2 km2. Observations of soil colour, depth, clast lithology and abundance were recorded at site. The samples were classified into five textural groups (sand, sand-silt, silt, silt-clay and clay).

The methods used for urban soils were similar except that (1) the sample density was higher, at four sites per square kilometre; (2) the sample sites corresponded closely to a predefined grid and did not avoid areas of human influence. In addition, extra samples requiring special treatment were taken for the determination of selected organic constituents (Smyth, 2009: especially Appendix 1).

At each soil sample site, information on the location, site and catchment geology, contamination, land use, and other features required for data interpretation were entered onto field cards. The sample location was also plotted on a field copy of the 1:50 000 Ordnance Survey of Northern Ireland (OSNI) map.

Observations from field cards were entered into a digital Access2000™ database after undergoing a field quality control process (Lister et al, 2005). This involved checking that the correct codes had been recorded on field cards and that GPS coordinates recorded on the card matched those in the GPS unit for each site. Thus both a traditional paper archive of observations was maintained as well as the construction of a computerised database.

Soils were initially air-dried at the field-base prior to transport to the sample store where they were dried in a temperature controlled oven at 30°C for 2–3 days. At the end of each field campaign samples were checked against field sheets prior to packing for transport to the BGS laboratory for sample preparation. On arrival at the laboratory samples were checked against shipping lists prior to assigning laboratory batch numbers in the BGS UKAS Quality Assurance System. The A and S soils were prepared in the same manner in a trace-level sample preparation laboratory.

Samples were disaggregated prior to sieving to a <2 mm fraction using nylon mesh. Replicate samples were prepared by riffle splitting each of the duplicate samples. Soil pH and LOI was determined for every A surface soil sample. A representative 30 g (± 2 g) sub-sample was obtained by cone and quartering. This sub-sample was then milled in an agate ball mill at 300 rpm for 30 minutes.

Different analytical procedures were employed for the surface and deep soils. Pressed pellet production and XRF analysis were completed by laboratory on surface soils only. Sub-samples of milled soil were weighed and placed into tamper-evident plastic sample tubes. The XRF pressed pellet was prepared by adding an aliquot (3 g ±0.05 g) of two blended synthetic waxes comprising 90 % EMU 120 FD wax and 10 % Ceridust (both waxes are styrene based co-polymers) to 12 g (± 0.05 g) of milled material. This mixture was milled for 4 minutes at 300 rpm. On completion of the binder milling the prepared powders were placed into tamper evident plastic sample tubes for temporary storage prior to pellet preparation. Pellets (40 mm) were pressed using a calibrated Herzog semi-automatic pellet press at 25 kN.

Prior to analysis, concealed certified reference materials and secondary reference materials were inserted into the sample batches. XRF analysis of the A samples was undertaken at the BGS; ICP analysis of A and S samples at SGS Laboratories, Toronto; and fire-assay of S samples at SGS Laboratories, Toronto.

For the Tellus samples, Energy Dispersive Polarised X-Ray Fluorescence (ED(P)- XRF) spectrometers were used to analyse those elements for which the WD-XRF spectrometers were insufficiently sensitive. Certified Reference Material (CRM) standards were used to calibrate the instruments. The PANalytical software was used for spectral deconvolution and to fit calibration curves, applying matrix correction by internal ratio Compton correction method. The calibrations were validated by analysis of a wide range of RMs. The detectors were calibrated weekly. All backgrounds and peaks were corrected for instrument drift using two external ratio monitors, when required. Quality control was maintained by regular analysis of two glass monitor samples containing 47 elements at nominally 30 mg/ kg and 300 mg/kg. Results were presented as run charts for statistical analysis using statistical process control software (SPC).

The lower limits of detection are theoretical values for the concentration equivalent to three standard deviations (99.7 % confidence interval) above the background count rate for the analyte in an iron-rich alumino-silicate matrix. For silicate matrices the practical detection limits for most elements approach the theoretical values due to high instrumental stability. LLDs were calculated from a matrix blank and the ‘synthetic’ Pro-Trace standards.

Individual results are not reliable below the quoted lower limits, but reliable estimates of the average or typical values over an area may be obtained at lower levels of concentration; meaningful distribution patterns may thus be recognised for some elements at levels lower than the LLD.

The data are described in Young, Mike; Donald, Alex, eds. 2013 A guide to the Tellus data. Belfast, UK, Geological Survey of Northern Ireland, 233pp. available for free download from: http://nora.nerc.ac.uk/509171/

The Tellus survey was funded by the Department of Enterprise, Trade and Investment (DETI), now the Department for the Economy (DfE) in Northern Ireland and the INTERREG IVA programme of the European Union (EU) Regional Development Fund.

This data was collected by the Geological Survey Ireland, the Department of Culture, Heritage and the Gaeltacht, the Discovery Programme, the Heritage Council, Transport Infrastructure Ireland, New York University, the Office of Public Works and Westmeath County Council. All data formats are provided as GeoTIFF rasters but are at different resolutions. Data resolution varies depending on survey requirements. Resolutions for each organisation are as follows: GSI – 1m DCHG/DP/HC - 0.13m, 0.14m, 1m NY – 1m TII – 2m OPW – 2m WMCC - 0.25m Both a DTM and DSM are raster data. Raster data is another name for gridded data. Raster data stores information in pixels (grid cells). Each raster grid makes up a matrix of cells (or pixels) organised into rows and columns. The grid cell size varies depending on the organisation that collected it. GSI data has a grid cell size of 1 meter by 1 meter. This means that each cell (pixel) represents an area of 1 meter squared. .hidden { display: none }