Resumen de Structure and physical properties of the subduction plate boundary
In the last 50 years, seismic methods have improved with the development of recording instrumentation, and even more in the last 20 years with increased computational power. This has permitted to model larger amounts of data and implementing increasingly accurate numerical models, which has in turn allowed to obtain better constrained, higher-resolution models of physical properties of the Earth’s subsurface. In particular, my PhD work focuses on the development and application of seismic tomography methods to improve seismic images and models of subduction zones and particularly the inter-plate boundary zone. It consists of two parts: the first part presents the application of joint refraction and reflection travel-time tomography (TTT) to model onshore/offshore wide-angle seismic (WAS) data recorded along a profile in the North Chilean subduction zone using the tomo2d code. This work allowed me to get knowledgeable with the subroutines of this code, an experience which was applied in the second part of my work. In this second part, I modified tomo2d to allow to jointly inverting WAS near-vertical multichannel sesimic reflection (MCS) and I showed the outstanding improvement of the inversion results by combining both types of data into a common inversion strategy.
In the field data application of the first part, I have applied TTT to WAS data recorded during the Crustal Investigations off- and on-shore Nazca/Central Andes (CINCA’95) survey along ~21°S in North Chile. Results show that there is a significant change in the overriding plate properties as well as the inter-plate boundary depth along the Chile Trench. The thickness of the plate increases northwards with the deepening of the inter-plate boundary at similar distances from the trench axis. Furthermore, there is a change in surface deformation of the overriding plate visible in the orientation of faults cutting the onshore relief. Fault orientation changes from trench-parallel in the south to trench-perpendicular in the north. The borderline between the areas of different fault orientation support that there is a change in the regional stress of the overriding plate along the North Chile convergent margin.
However, this application also revealed some weaknesses of the method. The main is that the poor ray coverage and limited redundancy of inter-plate reflections do not allow to recover the inter-plate boundary geometry and the properties around it with sufficient detail. I hypothesize that this problem can be overcome by combining different types of data containing different seismic phases into a common inversion. In particular I propose to combine near-vertical MCS data with WAS data, because these data sets are highly complementary in terms of azimuthal coverage and redundancy, so each one allows to overcome the limitations of the other. I show with synthetic tests that the velocity-depth ambiguity that is intrinsic to streamer MCS acquisition is compensated by the introduction of refraction travel-times from WAS and downward continued MCS data. Furthermore, the sparse WAS data coverage is compensated by high MCS data redundancy.
The tests show that the combination of WAS data with a small amount of MCS data produce highly improved velocity and reflector geometry models in terms of accuracy and resolution. The velocity models obtained using the proposed inversion scheme are more accurate that those obtained from WAS data alone, whereas the geometry of the inverted reflector is similar to that in MCS images. In particular, tests show that including 5-10% of the MCS travel-times into the joint inversion is enough to have optimal results. The TTT inversion of combined WAS and MCS data allows not only the improvement of velocity model and reflector depth, but it also minimizes the velocity-depth trade-off and thus provides better-constrained models that are easier to interpret based on the MCS images. We propose this strategy as the way to follow for the near future in the exploration of the inter-plate boundary of subduction zones.
