Estimating Local and Regional Spatial Variability of Soil Geochemistry Using Geostatistics for Environmental and Forensics Applications.

Information collected by ground and remotely sensed surveys can be useful for a range of applications including geological and soil mapping, environmental studies, estimation of elemental soil concentrations for health implications and prediction of the soil provenance for forensic purposes. These national and regional ground and remotely sensed surveys generate large scale spatial databases and coverage maps of soil properties such as soil geochemistry. An implicit assumption in these applications is that properties measured in the ground and remote sensed surveys are directly related to different aspects of the region such as underlying geology, superficial deposits and soil types. Assessing the characteristics and spatial variability of the variables measured in the ground or remotely sensed surveys provides a baseline for estimating regional variability. The spatial variability or continuity depends on the distribution of the soil property. The assumption may be that spatial variability can be assumed to be similar within similar parent rock or soil types depending on the level of discontinuities such as faults within a certain area. Geostatistical techniques provide the means to investigate and quantify spatial correlation. This study uses the core tool in geostatistics, the variogram, to estimate local scale spatial variability in soil geochemistry data and the spatial correlation with geology and superficial deposits. The data used were generated by a regional ground-based soil geochemistry generated as part of the Tellus project, Geological Survey Northern Ireland. The Tellus project was funded by the Northern Ireland Department of Enterprise Trade and Investment and by the Rural Development Programme through the Northern Ireland Programme for Building Sustainable Prosperity.