【摘要】 首先介绍方差分析叠加周期外推法,并以郯庐地震带公元1400年以来40次MS5级以上地震为例说明方差分析叠加周期外推法的应用方法,应用该方法分析了不同区域、不同震级的地震序列。结果显示:郯庐断裂地震带、北京-渤海地震带、苏鲁皖交界地区、安徽地区的地震存在显著周期,从统计学的角度可以预测下一次地震的范围。而河北平原地震带、淮河地震带、华东地区的地震经度或纬度则不存在显著周期,无法外推下一次地震的范围。方差分析叠加周期外推法用于地震预测尚属于一种尝试。外推预测值与原始实测值虽然存在一定的误差,但方法的意义在于应用统计理论上把预测地点进行了定量化,缩小了监视范围。从某种意义上,还为研究地震机理,探索地震在时、空、强的演变规律上,提供研究方向。更多还原

【Abstract】 This paper firstly introduces variance analysis superimposed by periodic extrapolation and takes 40 earthquakes with MS≥5 in Tanlu seismic belt since 1400 B.C as an example to illustrate the application of variance analysis superimposed by periodic extrapolation. The paper analyzes the seismic sequences of different regions, different earthquake magnitude by using this method. The results show that significant earthquake cycle exists in Tanlu fault earthquake zone, Beijing-Bohai seismic belt, Jiangsu-Shanghai-Anhui border area and Anhui region. From a statistical point of view, we can predict the next earthquake. But there’s no significant period in Hebei plain seismic belt, Huaihe seismic zone, and earthquake longitude or latitude in eastern China,which cannot extrapolated for the range of next quake.Variance analysis superimposed periodic extrapolation for earthquake prediction is still an attempt. In addition, the article also pointes out that there may be some errorsbetween extrapolation forecast value and measured value. But the significance lies in the application of statistical theory to predict the locations quantitatively, and narrowing the monitoring range. In a sense, it also provides research directions for the study of mechanism of earthquake, the exploration of earthquake rules in time, space and strength revolution.更多还原