S.M.A. Shah1
1Command and Staff College, National Defence University (NDU), Pakistan
Abstract:
Pattern Informatics can be used to predict seismic events. Three forms of pattern can be formed using global seismological data. Temporal patterns indicating a temporal relationship between different natural seismic events that took place over past years over at least two decades depict their sequence in the form of a pattern. Spatial patterns indicate the geographic locations of events and order of their occurrence. Most seismic events take place around active fault lines. However, structured study of the database is likely to reveal additional conclusions. Spectral patterns indicating the frequency characteristics of the events establish forms of interpretable correlations. A three dimensional matrix comprising all three patterns is likely to yield valuable results. However, as estimated, complete results strongly depend upon the completeness of the acquired data from around the globe used together. Variations in sensors and estimation techniques can always be catered for using normalization. A test case based on data collected and structured on the basis of regions, where possible, may also help establishing relationships. However, the importance of a global super set remains crucial and is called for. Simulations of available data yielded relevant and usable outcomes indicating possible future events with related scale and depth parameters.
