Modeling And Analysis Of Spatio-temporal Data Of Earthquakes Based On GIS And Complex Network Theory

Earthquake is one of the most serious and far-reaching geological disasters in the world.There is no specific rule for the seismic phenomena in a short period of time,but through the statistical analysis of a large number of seismic data,we can study the dynamic changes of seismic phenomena in a specific period of time.Therefore,it is urgent to study the law of earthquake activity and provide theoretical support for earthquake prevention and mitigation.However,due to the complexity and randomness of seismic activity,the inaccuracy of geophysical model and the limitation of people's understanding of crustal movement,earthquake prediction has become one of the recognized international problems.Previous studies have shown that seismic data have complexity,inheritance,hierarchy and periodicity in temporal and spatial distribution.Therefore,analyzing the evolution characteristics of time-space sequence of seismic data is an effective way to obtain the laws of seismic motion.GIS spatial analysis method is a kind of theory method which can deal with spatial data information effectively and has been widely used in the field of earthquake prevention and disaster reduction.Complex network theory provides a powerful theoretical and technical support for describing the internal structure,interaction mechanism and dynamic evolution of real complex system.It is believed that a large number of seismic events show certain statistical laws,and there are complex relationships and interactions between seismic activities.From the perspective of complex systems,the study of spatiotemporal data of earthquakes has become a new starting point for the study of seismic activities,and has been widely concerned by scholars at home and abroad.How to break through the traditional research ideas and analyze the dynamic characteristics of seismic spatiotemporal data from the perspective of spatial network has become a hot issue in seismic research.Therefore,this paper focuses on the temporal and spatial statistical characteristics of seismic sequence data.Combined with the spatial analysis method of GIS and the theory of complex network,the spatiotemporal network modeling of seismic data in Sichuan and Chongqing area is carried out,and some important statistical characteristics are analyzed.At the same time,the spatial aggregation behavior of seismic sequence data is studied,and the spatiotemporal analysis of seismic data is carried out from multiple perspectives.Specifically,the research work and achievements of this paper mainly include the following aspects:(1)The spatial characteristics of seismic data in the study area are analyzed from the perspective of spatial statistics and density by using GIS spatial analysis method.It is found that there is a linear relationship between the logarithm of the earthquake frequency and the magnitude of the earthquake;there is a power law relationship between the number of regions with the same earthquake frequency and their frequencies and a heavy tail feature;and the relationship between the average maximum magnitude of the region and the frequency of the region follows the law of allometric growth scale.(2)Considering the magnitude,frequency,location and time of the earthquake events in Sichuan and Chongqing area,this paper makes a statistical analysis of the earthquake point cloud in the study area.It is found that the spatial distribution of the earthquake points in Sichuan and Chongqing area is mainly concentrated on the Longmenshan fault zone,which has a very significant spatial clustering characteristic,and the earthquake frequency will increase sharply after a strong earthquake.(3)A method of seismic spatio-temporal network modeling based on stratigraphic lithologic data is proposed.The whole study area is divided into different regions by lithologic data,and then the seismic spatio-temporal network model of the study area is established according to the spatial relationship between seismic sequences and regions and the continuity of seismic sequences in time.The results show that the degree distribution and the arc weight distribution of the seismic time-space network show significant power-law characteristics.(4)In this paper,multiple weighted seismic time-space networks based on different time periods are constructed,and the statistical indexes of each network are calculated.The experimental results show that the degree distribution and the weight distribution pattern of the arc segment in the seismic network follow the power-law distribution,and the interaction between the active rock masses will change with time.In addition,the study on the changes of network average degree,average clustering coefficient and power-law index with time shows that most of the earthquakes occurred in the rock mass where earthquakes have occurred in the past,but the basic structural characteristics of the seismic network remain basically unchanged in the study area as time goes on.