Image Information (pi) Study In The Continental Parts Of The Application

Pattern informatics (PI) method was first introduced by Rundle et al. at 2000. It is a new way to forecast earthquake based on statistical physics. The fundamental principle of PI method is given as follows:The region of interest is divided into many square boxes. For each box, there is a time series, which is the instantaneous number of earthquakes per unit time. Then, the abnormal change of seismic active intensity was extracted from each box as well as the changes of seismic active intensity were normalized. After that, the probability of occurrences of significant earthquakes in each box was computed and then subtracts their corresponding background probability. At last, we extract these boxes (namely seismic hotspots) which have high probability of earthquake occurrence.In general, The PI method works for long-term earthquake forecasting (about 5～10 years), but it can give the highest spatial resolution of abnormal regions (for example, the spatial resolution of abnormal regions is 11km×11km, which is obtained by PI method for earthquake with M = 5). The PI method has been applied to sourthern Califrnia, Japan, Taiwan, Sichuan-Yunnan region, and other regions. The research results imply that the PI method is more effective than random forecast.Based on the sufficient investigation about the development history, the theory, the algorithm of PI method, we have written the code of PI method and applied the program to study mid-long term seismic risk assessment about Yunnan, north China, and other regions. In the study, the research approaches are given as follows:Firstly, we choose initial values to some parameters, such as the range of study region, the initial time of seismic data, the magnitude lower limit and so on. Secondly, we determined the values of other parameters, such as the forecast interval, the scale of box, the threshold and so on, by multiple retrospective testing. Thirdly, we singled out the optimum parameter values for the region of interest. What's more, we studied earthquake risk assessment in the region of interest in future, based on the optimum parameters.At last, the research results are summried. And, the question about parameters selection and other questions are further discussed and explained in this paper. Because the PI method is a new way to forecast earthquake, there are some limitations in this paper, and we will research and improve it in the future.