摘要
地质、物化探数据的分布常出现一个长尾巴,这种分布特征影响了矿产储量计算及数据处理的精度.引起这种非正态分布的因素有:数据中有特异值,它的出现表明矿床具有高品位的矿化作用,这对储量计算及数据处理是很重要的;数据由若干个总体组成,即所谓的混合总体.对这种数据,普通克立格法是不稳健的.为此,本文介绍一种新的地质统计学方法——指示克立格法(非参数方法).文中讨论了(1)指示函数及其二阶矩;(2)指示半变异函数;(3)指示克立格方程组;(4)待估块段品位的指示克立格估计.最后,还介绍1个应用实例,并给出了计算方法和步骤.
Geological or geophysical-geochemical exploration data curves usually show a long-tail of the distribution. Such a non-normal distribution feature will have an influence upon the precision of ore reservescalculation and data processing, being caused by following reasons: (1) the appearence of peculiar valuesin the data, showing a high grade mineralization of o the ore deposit dealt with and being of great importance to ore reserve calculation and data processing; (2) data with a multiple (or mixed) population.The universal kriging is not suitable for such data processing to get a steady solution. For this reason,a new geostatistical method-indicator kriging (a non-parametric method) is introduced in this paper, inwhich four questions arc discussed: (1) indicator function and its second order moments; (2) indicatorsemi-variogram; (3) system of indicator kriging equations; and (4) indicator kriging estimation of thegrade of the ore blocks to be estimated, including an estimation of the co-existing populations. In the final part of the paper an example of ore reserve calculation for a gold deposit is cited, showing the mathod and steps for the calculation.
出处
《地质与勘探》
CAS
CSCD
北大核心
1990年第3期28-36,共9页
Geology and Exploration
