Item

Abstract

Processing of seismological data requires knowledge of appropriate velocity model. Due to ongoing investigations of prominent geological regions usually numerous velocity models by different authors have been developed there. The question which arises is which one to use. This is also the case for Reykjanes Peninsula (RP) in SW Iceland, where the REYKJANET seismic network was built to monitor local seismicity. At present, five velocity models come into consideration in RP (four 1D gradient models and, one 3D tomographic model). We have tested all the five models by common methodology. It includes discussion of results of (i) hypocenter localization, (ii) P-wave polarization at station positions and, (iii) Rayleigh-wave dispersion. Alternatively, 1D velocity models have been combined with stations corrections thus yielding efficiently pseudo-3D models. All the tested models provide hypocenters whose traveltime residuals can be classified into two groups ((i) 1D models without station corrections and (ii) 3D tomographic model and 1D models with station corrections). Hypocenter coordinates differ slightly according the used model. Bigger differences between the models are due to the predicted P-wave polarization and Rayleigh-wave dispersion. This issue is connected with bigger sensitivity of those parameters to the surface properties of the models. Our proposition is to select the best one 1D velocity model supplemented with station corrections and use it anywhere possible. Using of a more complex 3D model would be usually not highly beneficial. 3D tomographic model is, however, one of not-so-many tools how to discover geological features of the underground.