Publication Date:
2020-02-12
Description:
An energy-flux model (EFM) and a teleseismic fluctuation wavefield method (TFWM) have been applied to interpret the teleseismic P coda observed at three temporary and two permanent networks in northern and central Europe. The aim is to determine the small-scale random structure of the lithosphere below the receivers. Various subsets of these data have been exploited previously with one of the two methods. The main objectives here are: To compare the performance of both methods with synthetic data sets. To map the random structure of the lithosphere in terms of inverse scattering Q (Q−1s), correlation length a, RMS velocity fluctuation σ, thickness L of the scattering layer and autocorrelation function (ACF) using a combination of both methods. With TFWM, the product aσ2 can be reliably determined if L is known. L can be roughly estimated with EFM. Although EFM can, in principle, resolve a and σ separately, a is recovered with a rather large uncertainty. TFWM does not distinguish much between the ACF type, whereas with EFM determination of the ACF type is sometimes possible. By combining the results of both methods we determine improved random medium parameters of the lithosphere for eleven subregions in northern and central Europe. In the Baltic shield, Rhenohercynian belt, Ardenne and Brabant mountains, eastern Rhenish massif, Eifel, Hunsrück mountains, Lorraine, Frankonian Jura and massif Central scattering predominantly occurs in the crust. For the frequency range from 0.5 to 5 Hz correlation lengths of 1–7 km and rms velocity fluctuations of 3-7 per cent are obtained. For the Rhenohercynian belt (RH) and the N-German basin RMS velocity fluctuations and correlation lengths could not be resolved. The data from the N-German basin (NB) cannot be explained by scattering within the crust only. Smallest scattering Q was found in the N German basin (Qs≈ 100 at 1 Hz) and largest scattering Q in the Baltic shield (Qs≈ 450 at 2-3 Hz). For the Frankonian Jura only a Gaussian ACF can fit the Q−1s values. The data from the eastern Rhenish massif also indicate a random medium with Gaussian ACF. For all other subregions we cannot distinguish between exponential or Gaussian ACF for the random medium structure.
Language:
English
Type:
info:eu-repo/semantics/article
Permalink