论文部分内容阅读
马特隆(Matheron)的理论带有狂热的性质。其早期信奉者竟打出了应用统计学专利的招牌,将其涂以些许概率和统计学的色彩并冠之以“地质统计学”的名字,于是这些人也就摇身一变而成了“地质统计学家”。遗憾的是,这种地质统计学理论与其老祖宗——数学概率和应用统计学相去甚远,而且终于坠入了克立格法的陷井。克立格方差夸大了各数据测点之间矿化作用的连续性,因此不能提供精度真实而又可靠的矿石储量估算值。品位方格划分法(grade-squared partition technique)将马特龙的基本理论与常规的块段设计以及各种已证实的统计学方法融为一体。这种方法不但能提供精度真实而又可靠的估算值,而且还是优化勘查设计的一种有效工具。
Matheron’s theory is fanatical. Its early followers hit the signboard of Applied Statistics Patents, colored it with a little probability and statistical color and branded “Geostatistics,” and these people were transformed into “Geostatistics Family”. Unfortunately, this theory of geostatistics is far from its ancestors - mathematical probability and applied statistics, and has finally fallen into the trap of kriging. Kriging variance exaggerates the continuity of mineralization between data points and therefore does not provide an accurate and reliable estimate of ore reserves. The grade-squared partition technique combines the basic theory of Matlon with the conventional block design and various proven statistical methods. Not only does this approach provide true and reliable estimates, but it is also an effective tool for optimizing survey design.