Earthquake catalogs provide a multitude of information for many fields of research. One of the key parameters of a catalog is its recording completeness. Completeness varies over space and time and its knowledge is essential for avoiding errors in statistical analyses of catalogs. Many properties, such as the b-value of the Gutenberg-Richter distribution or seismicity rates, depend strongly on recording completeness. Currently applied methods for estimating recording completeness are seismicity-based and determine completeness from earthquakes samples, which can raise many problems. Completeness values are only averages over space and time according to the sampling, no information is available in low-seismicity areas, and completeness is not independent of the Gutenberg-Richter distribution.
The new PMC-method (Probabilistic Magnitude of Completeness), developed by Schorlemmer and others, overcomes these problems of completeness estimates. It is based solely on empirical data: stationlists, phase-data and attenuation relations. From this data, probability distributions for each station in the network are generated, describing probabilities to detect earthquakes of particular magnitudes at particular distances from the station. Full probabilistic descriptions in space and time of the recording completeness of the network are computed combining these probability distributions. Thus, completeness becomes a function of the network and its stations instead of earthquake samples.
The proposed research shifts paradigms in estimating the magnitude of completeness of seismic networks and provides a comprehensive toolkit for scientists. This study, combining earthquake statistics, tomography, and wave form analysis, will establish a new and more encompassing seismic method.
All completeness estimates will be compiled into databases for distribution to the community. Besides improved completeness estimates, the work will offer a performance measure for network operators and managers.
