The high-frequency decay term of the acceleration spectrum κ is a commonly used parameter in engineering seismology. In recent years, the assumption of a linearly decaying spectrum in log–linear space has been recognized to not always be valid as the value of κ depends on the analyzed frequency band. We present an alternative model for the spectral falloff in which the frequency dependence is explicitly taken into account. This is motivated by observations that the quality factor Q has a power-law dependence on frequency at high frequencies. The new model describes the spectral decay with the help of two variables, opposite to the single parameter κ. The approach is applied to borehole data of the EUROSEISTEST site in Greece. The misfit between modeled and observed spectra is reduced with the new approach compared with the classical kappa model. The new estimates compare well with κ estimates if the same frequency interval is considered but additionally allows for the capture of the frequency dependence of the spectral shape.

Far away landslide detection A mass wasting and flood event on 7 February 2021 in Uttarakhand, India, killed more than 200 people and damaged two hydropower plants. Cook et al . discovered that teleseimic signals from the beginning of this event were recorded at different stations on a regional seismic network in northern India. The signals were observed up to 100 kilometers from the disaster and demonstrate the potential for these far-away monitoring stations to be useful for early warning. This discovery suggests a different way to monitor such remote Himalayan valleys for mass wasting hazards. —BG

S U M M A R YSituated in an active tectonic region, Santiago de Chile, the country's capital with more than six million inhabitants, faces tremendous earthquake risk. Macroseismic data for the 1985 Valparaiso event show large variations in the distribution of damage to buildings within short distances, indicating strong effects of local sediments on ground motion. Therefore, a temporary seismic network was installed in the urban area for recording earthquake activity and a study was carried out aiming to estimate site amplification derived from horizontal-tovertical (H/V ) spectral ratios from earthquake data (EHV) and ambient noise (NHV), as well as using the standard spectral ratio (SSR) technique with a nearby reference station located on igneous rock. The results lead to the following conclusions:(1) The analysis of earthquake data shows significant dependence on the local geological structure with respect to amplitude and duration.(2) An amplification of ground motion at frequencies higher than the fundamental one can be found. This amplification would not be found when looking at NHV ratios alone.(3) The analysis of NHV spectral ratios shows that they can only provide a lower bound in amplitude for site amplification.(4) P-wave site responses always show lower amplitudes than those derived by S waves, and sometimes even fail to provide some frequencies of amplification.(5) No variability in terms of time and amplitude is observed in the analysis of the H/V ratio of noise.(6) Due to the geological conditions in some parts of the investigated area, the fundamental resonance frequency of a site is difficult to estimate following standard criteria proposed by the SESAME consortium, suggesting that these are too restrictive under certain circumstances.

The one-dimensional (1-D) approach is still the dominant method to incorporate site effects in engineering applications. To bridge the 1-D to multidimensional site response analysis, we develop quantitative criteria and a reproducible method to identify KiK-net sites with significant deviations from 1-D behavior. We found that 158 out of 354 show two-dimensional (2-D) and three-dimensional (3-D) effects, extending the resonance toward shorter periods at which 2-D or 3-D site effects exceed those of the classic 1-D configurations and imposing an additional amplification to that caused by the impedance contrast alone. Such 2-D and 3-D effects go along with a large within-station ground motion variability. Remarkably, these effects are found to be more pronounced for small impedance contrasts. While it is hardly possible to identify common features in ground motion behavior for stations with similar topography typologies, it is not over-conservative to apply a safety factor to account for 2-D and 3-D site effects in ground motion modeling.

No abstract