A NEW FRACTAL APPROACH TO THE CLUSTERING OF EARTHQUAKES - PHYSICAL FRACTAL

Traditional fractal analysis methods of earthquake temporal clustering treat earthquake events as mathematical points (without mass) in time (Smalley et al., 1987). The results obtained by these methods mainly reflect the seismic behavior of numerous small earthquakes. A sieve me thod of physical fractal analysis has been developed in the present ar ticle in which a series of magnitude sieves are used to study the frac tal behavior of earthquakes with different magnitudes. Fractal analyse s of earthquake clustering are conducted for three regions: northern C hina, mainland of China, and southern California. We find that all the earthquake subsets of each region show nearly the same fractal dimens ion, D = 0.3 for northern China and the mainland of China and D = 0.4 for southern California. The scaling ranges in which the corresponding earthquake subsets under consideration exhibit fractal features vary from subset to subset. The method has the potential application to cha racterizing the fractal behavior of earthquake clustering for differen t regions.