试论克里金估计与随机模拟的本质区别

On the Essential Differences Between Kriging Estimation and Stochastic Simulation

简要介绍了地质统计学两大组成部分，克里金估计理论和随机模拟理论．克里金估计理论实际上是一种在均方损失下最优线性无偏估计，在应用上具有平滑效应，适合于描述变化不剧烈的连续地质变量的空间预测．随机模拟技术作为地质统计学的新领域，是当今储层表征技术一个重要组成部分，特别适合于对非均质性严重储层的刻画和描述，它产生各种等可能的储层随机图象，这一系列实现反映了储层特征空间分布的非均质性和不确定性．

Kriging estimation and stochastic simulation which are the two parts of geostatistics are introduced in this paper.Kriging estimation is the best linear unbiased estimation under the conditions of average square loss function.It has a smoothing effection and is adaptable for depicting geologic phynomenon with gradual variation.Being a new region of geostatistics,stochastic simulation is an important part of reservoir characterization.It has more advantages of quntifying reservoir with severe heterogeneities and various kinds of prbable stochastic images of reservoir.These realizations show the heterogeneities and uncertainties of spatial distribution of reserooir characteristics. The essential diffences between Kriging estimation and stochastic simulation are discussed.