We use data from 50 earthquakes in southern California to test the accuracy of event parameter determination using single seismic stations for the purpose of early warning. Earthquake magnitude, hypocentral distance, and backazimuth are all estimated using P-wave arrivals only. There is a wide range in the accuracy of event parameters determined by different seismic stations. One quarter of the stations produced magnitude estimates with errors less than 3.0ע magnitude units, hypocentral distances within 51ע km, and backazimuth calculations within 02עЊ. This accuracy is sufficient to provide useful early warning. Using P-wave arrivals is the most rapid method of delivering earthquake early warning and may permit a few seconds notice of impending ground motion even in the epicentral region. Our results show that networks using a P-wave detection approach for early warning can increase the accuracy of magnitude estimations by determining station-specific scaling relations between the predominant period of the P wave and event magnitude and by utilizing stations with optimal relations. Further, because individual stations are able to deliver an accurate early warning, the option of utilizing the technology in regions that lack a dense seismic network but are in need of seismic hazard mitigation becomes possible.
Scaling relations between the predominant period of P-wave arrivals and earthquake magnitude are explored using datasets from the Pacific Northwest and Japan, and compared with previous observations in southern California (Allen and Kanamori, 2003). We find the same scaling for events in all three geologically diverse regions. The sensitivity of the predominant period observation to magnitude can be optimized using various frequency bands for different magnitude ranges and in different regions. The ability to estimate the magnitude using the first few seconds of the P wave offers a methodology for earthquake early warning. The accuracy of magnitude estimates increases with the number of stations reporting predominant period observations. The most significant improvements in the magnitude estimate occur when the number of reporting stations increased from one to four. As in southern California, we find that the average absolute magnitude error is ϳ0.5 magnitude units when the closest four stations to the epicenter are used.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.