Astronomy and Astrophysics – Astronomy
Scientific paper
Jun 2003
adsabs.harvard.edu/cgi-bin/nph-data_query?bibcode=2003geoji.153..627s&link_type=abstract
Geophysical Journal International, Volume 153, Issue 3, pp. 627-644.
Astronomy and Astrophysics
Astronomy
1
Inverse Problems, Onikobe Earthquake, Seismic Tomography Method
Scientific paper
The consecutive subtraction of selected anomalies (CSSA) method has been presented to provide a new opportunity to improve the quality of inversion results in seismic tomography. The basic idea of the iterative method is to determine in each step a parameter that subtracts the total largest square residuals. The criteria proposed control the selection of the best solution from all candidates by inspecting their convergence to the accurate solution in the lsq sense. Resolution measures were defined by estimating the convergence degree. The matching of the Lanczos (LSQR) and CSSA methods has been demonstrated from the perspective of the projection technique. The storage and work requirements for both algorithms were compared, revealing that CSSA is also efficient for large and sparse systems. Synthetic models were reconstructed comparing CSSA and LSQR and indicating that CSSA gives us more reliable results, when large-sized structures are considered. At the same time, the simultaneous subtraction of the selected anomalies algorithm was examined. Numeral tests revealed that this algorithm is able to produce results that are similar to those obtained by LSQR. Complicated velocity structure can lead to ambiguous results, a problem that can be solved by setting the appropriate starting vector for the CSSA inversion. The difference between the starting vector and the starting (or reference) model is discussed. The CSSA method was applied to P-wave arrival times from shallow earthquakes to determine deep structure in the focal area of the Onikobe M= 5.9 earthquake (1996) in the northern part of Honshu (Japan). A specific feature of the traveltime residuals is a set of positive values corresponding to late P-wave arrivals. Tomography revealed a low-velocity body in the focal area from the surface down to the hypocentres. Local anomalies with the lowest perturbations were found in the surface layer of 0-2 km around the Onikobe epicentre, to the north of the focal area and to the northeast of Naruko volcano.
Hasegawa Akira
Horiuchi Shunsaku
Nikolaev Aleksei V.
Smaglichenko T. A.
No associations
LandOfFree
The method for consecutive subtraction of selected anomalies: the estimated crustal velocity structure in the 1996 Onikobe (M= 5.9) earthquake area, northeastern Japan does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.
If you have personal experience with The method for consecutive subtraction of selected anomalies: the estimated crustal velocity structure in the 1996 Onikobe (M= 5.9) earthquake area, northeastern Japan, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and The method for consecutive subtraction of selected anomalies: the estimated crustal velocity structure in the 1996 Onikobe (M= 5.9) earthquake area, northeastern Japan will most certainly appreciate the feedback.
Profile ID: LFWR-SCP-O-1821984