Abstract: In order to study the time-space complexity of seismic activity， in view of the difference in the interaction between seismic activity regions， an earthquake weighted network construction method based on time-space influence domain was put forward. California seismic data in 1984~2014 was selected to structure the earthquake weighted network， and its basic properties and connection properties of the network nodes were analyzed to reveal the inner link of earthquake activity. Experiments show that the link-weight distribution and the node-weight distribution can be described by power law; the relationship of node-weight and node-degree show that， for the property of this earthquack weighted network， the moderate-strong shock is as important as the large shock; by the study of the average neighbor degree and the clustering coefficient of seismic weighted network， it was found that strong shock regions are tend to be linked to strong connection. The above results show that the weighted network can help us find some internal relations of seismic system which cannot be found through the unweighted network.

Key words: time-space influence domain;earthquake networks, weighted network, connection correlation, scale-free