COUNTING STATISTICS OF F(-BETA) FLUCTUATIONS - A NEW METHOD FOR ANALYSIS OF EARTHQUAKE DATA

In a variety of biological and physical phenomena, temporal fluctuatio ns are found, which are not explainable as consequences of statistical ly independent random events. If these fluctuations are characterized by a power spectrum density S(f) decaying as f(-beta) at low frequenci es, this behaviour is called 1/f noise. Counting statistics applied to earthquake activity data leads to three time scales with different ch aracteristics, represented by the exponent beta: at interval lengths l ess than 1 h, the shocks are randomly distributed as in a Poisson proc ess. For medium time intervals (1 day to 3 months), the exponent 1 + b eta is larger (1.4 for M(o) = 3), but approaches unity for higher thre shold magnitudes M(o). In longer time ranges the exponent assumes valu es near 1.55, however, with increasing statistical variation at higher M(o), due to lower counts. The temporal sequence is different from wh ite noise; thus, it might be fruitful to apply neural network algorith ms, because this method allows predictions in some other cases with si milar characteristics.