贝叶斯最大熵地统计学方法及其在土壤和环境科学上的应用

贝叶斯最大熵(Bayesian Maximum Entropy,BME)地统计学方法是近年来出现的一种时空地统计学新方法。相对于传统的克里金方法,该法具有坚实的认识论框架和方法学基础。它不需要作线性估值、空间匀质和正态分布的假设,能够融入先验知识和软数据,并且不会损失其中蕴含的有用信息,提高了分析精度。本文首先介绍了BME的基本理论及其估值方法,随后简单描述了该方法的理论发展过程及其在土壤和环境科学上的应用情况,最后对该方法的应用做了总结与展望。经过国外研究者多年的开发和实践,BME方法已经被证明是一个理论上较为成熟,能够应用到实际研究中的优秀地统计学方法,在资源环境评估上有着广泛的应用前景。

The Bayesian maximum entropy geostatistical approach and its application in soil and enviromental sciences

The Bayesian maximum entropy (BME) approach has emerged in recent years as a new spatio-temporal geostatistics methods. By capitalizing on various sources of information and data, BME introduces an epistemological framework which produces predictive maps that are more accurate and in many cases computationally more efficient than those derived with traditional techniques. It is a general approach that does not need to make assumptions regarding linear valuation, spatial homogeneity or normal distribution. BME can integrate a priori knowledge and soft data without losing any useful information they contain and improve accuracy of the analysis. This paper first introduces the basic theory of BME and stages of BME estimation, and then briefly describes its development and application in soil and environmental sciences. Finally the application of this method is also summarized and prospected. After years of development and practice, the BME method has been proved to be a mature outstanding approach, which has a broad prospect of application in evaluation of resources and environment.