5:15 PM - 7:15 PM
[SIT20-P05] Using the Sdiff phase in waveform inversion: Preliminary application to D″ beneath the central Pacific
Keywords:waveform inversion, D″, lowermost mantle, Pacific LLSVP
The D″ region is the thermal boundary layer at the lowermost several hundred kilometers of the mantle and plays an essential role in mantle dynamics. The two large low-shear-velocity provinces (LLSVPs) that are known to exist in D″—one beneath the Pacific and another beneath Africa—are expected to provide important implications on the dynamics and evolution of Earth’s mantle and have been extensively studied since their discovery in the 1980s. However, whether LLSVPs are thermal upwellings, slab graveyards, primordial material, or iron-rich anomalies resulting from core-mantle interaction is still unknown. Part of the difficulty in investigating the origin of LLSVPs comes from the lack of detailed tomographic studies due to sparse data coverage. While seismic phases observed at epicentral distances shorter than 100º, such as S and ScS, are mainly used to infer structure in D″, the area within the LLSVP sampled by these phases is limited. Therefore, studies investigating the localized structure of LLSVPs have only been performed in several marginal areas of the LLSVPs. On the other hand, the Sdiff phase (S waves diffracted around the core-mantle boundary) samples larger areas of the LLSVPs and may be used to investigate these areas. One such area is the central Pacific, which corresponds to the northeastern edge of the Pacific LLSVP and where the plume beneath Hawaii is thought to originate.
In this study, we aim to develop a method for waveform inversion including the Sdiff phase in the data to infer the detailed 3-D structure in D″ beneath the central Pacific. While the Sdiff phase is known to exhibit the accumulated effect of structure across a large area along its raypath in D″, we test whether the use of a large amount of data with varying geometry will enable lateral heterogeneities to be resolved. We also test whether the vertical structure can be resolved from the differences in frequency response to anomalies. We use waveform inversion, which can extract information from the whole waveform, including postcursory signals. To overcome the issue of the Sdiff phase lacking a reference phase, we apply event and station corrections to eliminate the effect from the structure outside our target region. We compute synthetic waveforms for virtual structures and test whether the structure can be recovered using our methods. We will also show preliminary results of our inversion for the 3-D structure in D″ beneath the central Pacific, which may lead to a better understanding of the dynamics at the northeastern edge of the Pacific LLSVP.
In this study, we aim to develop a method for waveform inversion including the Sdiff phase in the data to infer the detailed 3-D structure in D″ beneath the central Pacific. While the Sdiff phase is known to exhibit the accumulated effect of structure across a large area along its raypath in D″, we test whether the use of a large amount of data with varying geometry will enable lateral heterogeneities to be resolved. We also test whether the vertical structure can be resolved from the differences in frequency response to anomalies. We use waveform inversion, which can extract information from the whole waveform, including postcursory signals. To overcome the issue of the Sdiff phase lacking a reference phase, we apply event and station corrections to eliminate the effect from the structure outside our target region. We compute synthetic waveforms for virtual structures and test whether the structure can be recovered using our methods. We will also show preliminary results of our inversion for the 3-D structure in D″ beneath the central Pacific, which may lead to a better understanding of the dynamics at the northeastern edge of the Pacific LLSVP.