基于马尔科夫链的裂缝期次分析——以重庆天府地区上二叠统长兴组露头为例

Analysis of Fracture Stages Based on Markov Chain:a Case Study of Changxing Formation of Upper Permian in Tianfu Area,Chongqing

为寻找一种经济又有效的裂缝期次定量划分方法,以重庆天府地区上二叠统长兴组地层为例,对构造裂缝形成期次的划分方法进行了研究。在大量野外观察与测量的基础上,以基于概率统计学的马尔科夫链为方法,进行了转移频数统计、马尔科夫性质检验、随机值计算、残差矩阵分析,得出研究区的裂缝发育期次主要为三期:第一期裂缝为近东西向,第二期裂缝为南东东向,第三期裂缝为北东向和南南东向;结合基于地球化学的充填物碳氧同位素分析,取对应期次的裂缝充填物进行实验,对上述分析结果进行佐证,结果发现其同位素值主要散落在三个区域:第一期裂缝充填物δ18O平均值为-8.58‰,δ13C平均值为3.32‰,形成时平均埋深为3864 m,形成温度为130℃;第二期裂缝充填物δ18O平均值为-7.16‰,δ13C平均值为3.79‰,形成时平均埋深为3084 m,形成温度为107℃;第三期裂缝充填物δ18O平均值为-5.18‰,δ13C平均值为3.70‰,形成时平均埋深为2147 m,形成温度为79℃。综合研究区构造演化史以及埋藏史分析认为:第一期裂缝形成于燕山早期,第二期裂缝形成于燕山晚期,第三期裂缝形成于喜山早期。实验结果成功验证了马尔科夫链的分析,证明马尔科夫链为一种可行的裂缝期次研究方法。

To find an economical and effective method to divide fracture stages quantitatively,the Upper Permian Changxing Formation in Tianfu area of Chongqing,SW China,was set as an example to study the division of tectonic fracturing.Based on a large number of field observation and measurement,the Markov chain,a method based on probability statistics,was used to perform transfer frequency statistics,Markov property test,random value calculation,and residual matrix analysis.Results show that three main stages of fracturing in the study area could be recognized:the first stage in EW orientation,the second stage in SEE,and the third stage in SSE and NE.Combined with the carbon and oxygen isotope analysis,the fracture fillings of the corresponding period were taken for experiments.Results show that the averageδ18 O value of the fillings of the first-stage fracture was-8.58‰,average value 3.32‰,average buried depth 3864 m,and the formation temperature 130°C.Those of the second stagewere-7.16‰,3.79‰,3084 m,and 107°C;and of the third stage,-5.18‰,3.70‰,2147 m,and 79°C,respectively.After analyzing the regional tectonic evolution and burial history,it was believed that the first stage fracture was formed in the early Yanshanian,the second stage fracture in the late Yanshanian,and the third stage fracture in the early Xishanian.The experimental results validate the Markov chain analysis successfully.Therefore,the Markov chain analysis is a feasible method for the study of fracture stages.