地理情報学と地球統計学: 概要

Application of Geostatistical Techniques for the Determining of an Anomalous Zone of Copper Ore Deposit in the Area of Polkowice Mine (Region of Lubin-Sieroszowice, SW Part of Poland)

Barbara Namyslowska-Wilczynska

This paper presents the results of investigations into the spatial variation of the principal geological parameters of strategic mineral resources, namely the copper ore deposits of Polkowice mine, occurring in the Lubin-Sieroszowice region (SW part of Poland). The studies were based on the data obtained by sampling of the Cu ore deposit with groove samples distributed mostly uniformly (at a spacing of 15-20 m) over the P-1 area of Polkowice mine. The studies concerned the Cu grade, thickness and quantity (accumulation) of the (recoverable) deposit concentrated in Whiteliegendes sandstones, Zechstein copper-bearing shales and calcareous-dolomitic formations. Selected geostatistical methods, i.e. ordinary kriging, bundled indicator kriging and conditional turning bands simulation have been used for the determining of an anomalous zones of Cu ore deposit.

In first stage of spatial studies ordinary (block) kriging for the estimation of the estimated averages Z* (with minimum variance of estimation σk 2) of deposit parameters within the P-1 block was used. Isotropic variograms were calculated and modelled by using of geostatistical models and then ordinary kriging was applied. Next, the indicator semivariograms for the following cutoff values of the studied parameters: Cu grade>1.7%, 2.2%, 2.7%; thickness>2.5
m, 3.0 m, 3.5 m and quantity>20[%mt/m2],>25 [%mt/m2],>30 [%mt/m2], were calculated. Probabilities P values in the block centres of the elementary grid, covering mining block P-1, were estimated by means of the bundled indicator kriging, taking into account the parameters of the indicator semivariograms models. As a result of the performed indicator estimations, the pictures of the probabilities P values distributions in raster maps of the exceeding of the assumed cutoffs were obtained.

In a successive stage of the analyses a conditional turning bands simulation technique was applied. The turning bands simulation was performed, taking into calculations of the theoretical models fitted to the Gaussian semivariograms of the deposit parameters. Statistical maps of the simulated values Zs distributions for quantity (mean of simulated values Zs, standard deviation of realizations σs, largest realization Zs, smallest realization Zs) and raster maps of probabilities P of exceeding of the assumed thresholds ( above cutoff values of quantity > 20[%mt/m2], >25[%mt/m2], > 30[%mt/m2] and below cutoff values < 20[%mt/m2], <25[%mt/m2], < 30[%mt/m2], were elaborated.
A comparison of the obtained results concerning used geostatistical techniques makes possible a better choice, indicating more useful and efficient method for estimating of mineral resources parameters and the determining of an anomalous zone of copper ore deposit.