【摘要】 <正> 在正常情况下,许多前兆观测资料可视为一个平稳的时间序列。在发生强震的时期,由于地壳及有关的地球物理场发生剧烈变化,这种平稳性才遭到破坏。问题在于如何从震前的、基本上是平稳的数据序列中找出统计性质的变化,预测强震的发生。对平稳过程而言,除常数均值外,自相关系数是其最重要的统计特征。计算相关系数,并考察它更多还原

【Abstract】 The 0-1 series method is adopted in calculating the autocorrelation coefficients of preeursory sequences. According to the processing result of the short leveling data at Zhangdaokou and Jinxian stations, it is found that the statistic characteristics reflected by the autocorrelation efficients are changing obviously in the period prior to the great Tangshan earthquake and the Haicheng event (May 18, 1978, MS=6.0). Finally, such variations of autocorrelation efficients are explained and discussed physically.更多还原