To retrieve the parameters of a seismic source from seismograms means to deconvolve the response of the medium from the seismic records. Thus, in general, source parameters are determined from both the seismograms and the Green function describing the properties of the medium in which the earthquake focus is buried. The quality of each of these two data sets is equally significant for a successful determination of the source characteristics. As a rule, both the sets are subject to contamination by effects, which decrease the resolution of the source parameters. Seismic records are usually contaminated by noise which appears as a spurious signal not related to the source. Error in the Green function is usually caused by the use of an improper model of the medium due to generally common poor knowledge of the seismic velocity of the area under study. Then, the phenomena of the structure which are not modelled in the Green function are assigned to the source where they distort the source mechanism.
To demonstrate these effects, we performed a synthetic case study simulating seismic observations in Dobra Voda locality at Male Karpaty region, Slovakia. A simplified 1-D and 3-D laterally inhomogeneous structural models were constructed, and the synthetic data were calculated in the 3-D model. We then used both the models during moment tensor inversion. The synthetic data were contaminated by a random noise up to 10 % and 20 % of maximum signal amplitude. We compared the influence of these two effects on retrieving moment tensors. It turned out that a poor structural model could be compensated by a high quality data, and, in similar way, a lack of data could be compensated by a detailed model of the medium. As an example, three local events of magnitude higher than 2 from Dobra Voda locality were processed.
The paper is available in PDF (586 kB).