【摘要】 利用滇西北地区震例资料和已有预测指标研究成果,采用层次分析法建立区域综合预测指标体系,获得了各预测指标的权重系数,在此基础上计算综合判定指标Y值并进行预测效能检验。结果显示:(1)本文建立的层次分析结构充分考虑了异常的共性和群体特征,得到了16项预测指标的相对重要性排序,权重值相对较大的指标分别是短期预测指标中的水化学突跳、前兆准同步、小震活动异常以及中期预测指标中的定点形变异常、流体破年变等;(2)Y值考虑了各级预测指标的权重系数,减少了由异常频次来分析震情的主观性,综合预测结果更加合理。震例检验显示综合判定指标Y值与主震震级存在一定的线性关系,且随主震震级的增大而逐渐增大;(3)滇西北地区Y值可以较好地反映中强地震的震兆强弱程度和异常可靠性,为区域震情综合判定提供定量化决策依据。更多还原

【Abstract】 In this article,we establish the prediction indexes system of moderately strong earthquakes in northwest Yunnan by using analytic hierarchy process(AHP),based on the earthquake cases in China and the research results of existing predictor indexes,and obtain the weight of each prediction index and the earthquake comprehensive judgment index Y,and tested the prediction efficiency of the comprehensive judgment index.The results show that:(1)With the fully consideration of the commonness and group characteristics of anomalies in the analytic hierarchy process structure we obtained the relative importance ranking of 16 prediction indicators.The prediction index with relatively large weight values are hydrochemical jump anomaly,precursory anomaly quasi-synchronization,small earthquake activity anomaly in the short-term prediction index,as well as fixed-point deformation anomaly and fluid annual change in the medium-term prediction index,etc.(2)The Y value takes into account the weight coefficients of the prediction indexes at all levels,reducing the subjectivity of analyzing the earthquake situation from the abnormal frequency,and making the comprehensive prediction results more reasonable.The earthquake case test shows that there exists a clear linear relationship between the comprehensive judgment index Y and the earthquake magnitude,which gradually increases with the increase of the magnitude.(3)Since the Y value in northwest Yunnan better reflects the earthquake precursor intensity and abnormal reliability of moderately strong earthquakes,it provides us a quantitative decision-making basis for the comprehensive analysis of regional earthquake situation.更多还原