Search

A new locally calibrated M L scale was derived for the area around the Shanxi rift system, North China, from 83 events recorded on 56 stations of the Shanxi Seismic Network (SSN) resulting in 2,633 observations during the period of 2008–2012. It is expressed as M L = log(A) + 0.80log(R) + 0.00187R − 1.4, where A is the maximum amplitude of vertical component in nanometer (nm) measured on a simulated Wood–Anderson seismogram at a hypocentral distance R. The new M L scale is valid for distances up to 600 km and is an improvement comparing to the currently used SSN scale, with a reduction in variance of 39 % in magnitude residuals. The moment magnitude M w was also calculated for the whole data set based on spectral analysis. To validate the Q-function used for spectral M w, we carried out moment tensor inversion for 17 moderate size events and selected the regional attenuation model which gave spectral M w on average closest to moment tensor M w. The most appropriate Q-function was found to be Q(f) = 299.4f 0.563. The determination of the M w magnitudes makes it possible to derive a relationship between the M L and M w scales of this region, which is expressed as M w = 0.85M L + 0.58, similar to relationships found elsewhere for similar tectonic environments.

In ground motion simulation studies based on earthquake rupture scenarios, the absolute ground motion level, frequency content, signal duration and distribution of simulated ground motion is highly dependent on the input parameters used in the calculations. We conduct a predictive study assessing the potential ground motions for Izmir, Turkey based on earthquake scenarios. We calculate ground motions from a reference scenario (Mw6.8) and compare it to 25 test scenarios for the Izmir fault in order to investigate the effect of input parameter uncertainty on the simulated ground motion. In this study, we use a hybrid broadband frequency (0.1–10 Hz) ground motion simulation technique. We find average ground motion levels in the central area of Izmir of more than 600 cm/s2 and 60 cm/s for peak ground acceleration and velocity, respectively, while the standard deviation is found to exceed 200 cm/s2 and 20 cm/s, respectively. The tested parameters identified to have the largest influence on the absolute ground motion level are seismic moment, average stress-drop, rise time and rupture velocity. Parameters identified to have the least effect on ground motion level are rake and ratio of stress-drop on asperities to background stress-drop. The low-frequency ground motion, critical in terms of damage to the building stock in Izmir, is mostly affected by the location of nucleation point, seismic moment, depth of the rupture area, rise time and the velocity model. It is, therefore, highly recommended that future studies focus on reducing the uncertainties in these input parameters.

Ground motion prediction equations (GMPE) in terms of macroseismic intensity are a prerequisite for intensity-based shake maps and seismic hazard assessment and have the advantage of direct relation to earthquake damage and good data availability also for historical events. In this study, we derive GMPE for macroseismic intensity for the Campania region in southern Italy. This region is highly exposed to the seismic hazard related to the high seismicity with moderate- to large-magnitude earthquakes in the Appenninic belt. The relations are based on physical considerations and are easy to implement for the user. The uncertainties in earthquake source parameters are accounted for through a Monte Carlo approach and results are compared to those obtained through a standard regression scheme. One relation takes into account the finite dimensions of the fault plane and describes the site intensity as a function of Joyner–Boore distance. Additionally, a relation describing the intensity as a function of epicentral distance is derived for implementation in cases where the dimensions of the fault plane are unknown. The relations are based on an extensive dataset of macroseismic intensities for large earthquakes in the Campania region and are valid in the magnitude range Mw = 6.3–7.0 for shallow crustal earthquakes. Results indicate that the uncertainties in earthquake source parameters are negligible in comparison to the spread in the intensity data. The GMPE provide a good overall fit to historical earthquakes in the region and can provide the intensities for a future earthquake within 1 intensity unit.

The seismically active Skagerrak region in the border area between Denmark and Norway has traditionally been associated with uncertain earthquake locations due to the limited station coverage in the region. A new seismic station in southern Norway and a recent update of the earthquake database of the Danish National Network have led to a much more complete and homogeneous data coverage of the Skagerrak area, giving the possibility of improved earthquake locations in the region. In this study, we relocate earthquakes in the Skagerrak area to obtain a more exact picture of the seismicity and investigate well-recorded events to determine the depth distribution. Hypocenter depths are found to be generally in the range 11–25 km. Furthermore, new composite focal mechanisms are determined for clusters of events with similar waveforms. Results indicate that the Skagerrak seismicity is associated with shallow, crustal faults oriented in the NS direction south of the Sorgenfrei–Tornquist Zone (STZ) as well as with the STZ itself. Mainly reverse faulting mechanisms along NE– SW oriented faults indicate maximum horizontal compression in the NW–SE direction. This is in agreement with World Stress Map generalizations, most likely associated with ridge push forces from the mid-Atlantic ridge, though modified probably by local crustal weaknesses. publishedVersion