NA-GCAP: A New Full Moment Tensor Inversion Method Based on the Renewed GRTM and NA Method

Abstract

Moment tensor inversion plays a crucial role in determining earthquake types, magnitude, and source geometry. Compared to polarity and amplitude methods, full-waveform approaches provide more comprehensive constraints on the complex full moment tensor (FMT). In this study, we propose a novel FMT inversion method named neighborhood algorithm-generalized cut-and-paste (NA-GCAP) method. Similar to the “cut-and-paste” method, our approach divides seismograms into Pnl and surface-wave segments. To enhance inversion efficiency, we employ the NA in conjunction with a newly developed strategy for efficient Green’s functions computation based on the renewed fast generalized reflection and transmission method (GRTM). Our method requires only a single forward computation and stores ten Green’s functions, significantly improving computational efficiency. We validate the robustness of our approach through synthetic tests using a velocity model perturbed by 5% relative to the input model. Furthermore, we apply the NA-GCAP method to the 2019 Changning earthquake sequence comprising 16 earthquakes, where twelve events exhibit double-couple (DC) components larger than 0.95, indicating a simple dislocation source, and four events display significant non-DC components. Our results align well with the previous studies and demonstrate the potential for widespread application to other earthquake sequences in the future.