 |
| Titel |
Source parameters for global seismology: how accurate are CMT-style source inversions? |
| VerfasserIn |
A. P. Valentine, J. Trampert, J. H. Woodhouse |
| Konferenz |
EGU General Assembly 2012
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 14 (2012) |
| Datensatznummer |
250062456
|
|
|
|
|
|
| Zusammenfassung |
| The centroid–moment-tensor (CMT) algorithm, introduced by Dziewonski et al. (1981),
has found wide application: the Global CMT Catalogue contains details of over
35,000 earthquakes, and is routinely used by researchers across the spectrum of
geosciences. However, comparatively little attention has been paid to the uncertainties
associated with the source parameters: what error-bars should be put on a CMT
solution?
We discuss the limitations and possible sources of uncertainty within the CMT
algorithm. In particular, we focus on the potential for systematic errors, arising from
the use of approximate forward-modelling techniques and low-resolution earth
models. By conducting experiments using synthetic data, we are able to map out the
misfit function that the CMT algorithm seeks to minimise. We observe that whilst a
minimum-misfit solution is typically well-defined, the curvature of the function
may be very low—so that in some cases, large perturbations in source parameters
result in only small increases in misfit. This implies that these parameters are poorly
constrained, and have large uncertainties that should be considered during subsequent
analysis.
For any specific seismic event, assessment of source parameter uncertainties requires
knowledge of the dataset and detailed inversion strategy used—information that is often
lacking when catalogue sources are used. Where the accuracy of source parameters is
particularly important, we suggest that users should consider performing their own
inversions. However, this may not always be practical; therefore, we show that it is possible
to estimate the misfit function curvature using only forward-modelling, and that this allows
some assessment of relative uncertainties to be made.
Dziewonski, A., Chou, T.-A. and Woodhouse, J., 1981. Determination of earthquake source parameters from
waveform data for studies of global and regional seismicity, Journal of Geophysical Research, 86, pp.2825–2852. |
|
|
|
|
|
|