Abstract:
Earthquake Alarms Systems, ElarmS, is designed with the goal of providing warning of forthcoming ground shaking during earthquakes. The event magnitude is found to scale with the maximum predominant period, τp max, which is estimated, using the frequency content of the first few seconds of the P-wave arrival from the vertical component of velocity records. I tested ElarmS offline using 242 earthquakes with magnitudes between 0.5 and 5.1 occurred across The Gulf of Gökova, Turkey. The test data for small magnitude events were obtained from the detailed re-processing of August 2007 with additional data from temporary stations. The larger events were directly taken from NEMC catalogue. I found that events smaller than 3.0 did not have clear relation between τp max and event magnitude, on the other hand, events with magnitude larger than 3.0 showed a scaling relation between τp max and event magnitude. The relation obtained for Gokova was consistent with the ones obtained in California and Japan. The average magnitude error between real local magnitude and ElarmS magnitude was about 0.91 magnitude units when single closest station to the epicenter was used. Once data from the two closest stations available the error drops to 0.62, when six station data available the error drops to 0.49.
