地形因子与小微地震相关度的空间异质性分析

针对传统最小二乘回归未能顾及数据的空间特性,且无法度量模型自变量与因变量相关性的空间变异特性的问题,本文提出利用地理加权回归方法分析小微地震频次与地形因子相关度的空间异质性。以四川地区的地震监测资料、DEM为实验数据,选取地形复杂度、坡度变率、坡向变率和地面曲率为自变量,地震发生频次为因变量,构建地理加权回归模型,并进行回归系数的空间变异分析。实验分析发现,地震频次与地形因子具有一定的相关性:地形复杂度与地震频次相关性最强;坡度变率、沟壑密度、剖面曲率与地震频次的相关性依次减弱;不同空间位置的地形因子和地震频次的相关性具有较明显的空间异质性。实验结果表明,地理加权回归可以有效地度量分析地震频次与地形因子相关度的空间异质性,研究结果可为地震及次生灾害的分析与预报提供辅助决策参考。

Spatial Heterogeneity Analysis of Correlation between Small Earthquakes and Topographic Factors

In order to solve the problem that the traditional least square regression fails to take into account the spatial characteristics of the data and cannot measure the spatial variability of the correlation between the independent variables and the dependent variables of the model,a geographically weighted regression method is proposed to analyze the spatial heterogeneity of the correlation between the frequency of small earthquakes and the topographic factors.Taking the earthquake monitoring data and DEM in Sichuan as experimental data,the terrain complexity,slope variability,slope aspect variability,and terrain curvature as independent variables,and the frequency of earthquake occurrence as dependent variables,a geographically weighted regression model was constructed and the spatial variation analysis of regression coefficient was carried out.It is found that there is a certain correlation between the earthquake frequency and the topographic factor in Sichuan.The correlation between terrain complexity and earthquake frequency was the strongest;The correlation among slope rate,slope aspect variability,terrain curvature and seismic frequency decreased in succession;The correlation between topographic factors and earthquake frequency in different spatial locations had obvious spatial heterogeneity.The experimental results show that geographically weighted regression can effectively measure the spatial heterogeneity of the correlation between seismic frequency and topographic factor.The results can be used as a reference for the analysis and prediction of earthquake and secondary disasters.