摘要
b值是从实际地震资料的统计中得到的.它与实际资料的完整性、可靠性、统计样本量的大小、取样的时空范围、样本的起始震级和取样间隔等因素都有关系. 本文使用累积频度-震级关系,lgN=a-bM,以起始震级为4 3/4和5;震级间隔为1/4,1/2和1级对不同尺度的时空范围进行统计,从中寻找各因素对b值统计结果的影响,为如何合理统计b值取得了一些实验依据. 此外,本文还就历史地震资料和现代仪器记录资料的统一使用问题,lgN-M线性关系两端“掉头”对b值统计的影响问题也作了探讨,从而提出在地震危险性分析中b值选取的原则及合理选取b值的方法.最后讨论了b值统计中的误差范围及其对危险性分析所产生的影响.
The b value is estimated from the statistics of actual earthquake data. It depends on the completeness and reliabilty of the data,the sampling size,sampling range in time and space,the threshold magnitude of earhquakes and the sampling interval,etc.In this paper,we use the relationship lgN = a-bM between the accumulative frequency and magnitude. In order to search for the after effects of various factors on the result of b value estimation,we estimate the b value with various thresholdmagnitude and magnitude interval for different temporal and spatial ranges respectively. We found some way for reasonably obtaining the b value with experimental basis.In addition,we discuss the problem of using historical data and recent observed data simultaneously and investigated also the influence of the 'bending' at both ends of the linear relationship,lgN = a - bM on b value estimation. Then we suggest the principle and method of choosing a reasonable b value in seismic hazard analysis. Finally,we discuss the error range in estimating the b value and its impact on hazard analysis.
出处
《地震学报》
CSCD
北大核心
1989年第4期351-361,共11页
Acta Seismologica Sinica
