|
Case histories |
Mineral Resource Evaluation Research Unit, Department of Geology, University of Wales, College of Cardiff, PO Box 914, Cardiff CF1 3YE, UK
In June 1990, a research project was commenced at a site in Southern England, in which sands and gravels of the Kesgrave Formation are overlain by Anglian boulder clay. This site was, at the time, under evaluation by RMC (UK) Ltd as a potential new source of natural aggregate.
The primary aim of the research was to ascertain whether standard geostatistical techniques could be applied to Quaternary fluvio-glacial sand and gravel deposits. Comparisons would then be possible with the results of evaluations carried out using inverse distance weighting and digital terrain modelling techniques.
The data used in the project were collected by shell and auger drilling on a localized detailed drilling grid (25 m spacing), and a less regular grid (100 m hole spacing) covering the whole site. The latter was carried out as part of a standard site evaluation. The variables used in the study included mineral and overburden thickness, size gradings as defined by BS:812, and petrography.
The work has highlighted the need for improved methods of drilling and sampling in unconsolidated deposits. Loss of fine (<75 µm) material and poor sampling of the coarser (> 40 mm) fraction had resulted in excessively high nugget variances which caused difficulties in the modelling of semi-variograms for these size fractions.
This paper presents the results of the geostatistical modelling of the above parameters using a range of methods including indicator variography and kriging. In particular, the deviation of the kriged block values from specific British Standards was assessed in order to determine the potential end use of the in situ material, before or after selective blending/processing.
