Variable Global Grid Refinement and Prediction Using RSTT Model
Description:
Regional seismic travel time (RSTT) is a global model (Myers et al. 2010, Begnaud et al. 2021) that rapidly predicts travel times of regional seismic phases (Pn, Sn, Pg and Lg), while accounting for key effects of the 3-dimentional crustal and upper mantle structures on the regional travel times. Previous versions of RSTT model utilizes, a model grid of constant one-degree cells and in a recent study Babikoff et al. (2022) focused on Israel and the Middle East, showed that iterative data driven grid refinement improves the resolution of P wave (Pn and Pg) tomography in the tectonically complex region. We are further refining the methods outlined by Babikoff et al. (2022) to improve the travel time tomography of regional seismic phases globally to explore the effects of model parameterization with the iteratively decreasing grid sizes as well as trade-offs relating to the retrieved crustal and upper mantle velocity structures.
We conducted tomography for upper mantle Pn velocity and Pn gradient iteratively on a global scale at grid resolutions of 1.0°, 0.5°, 0.25°, and 0.125°, systematically varying the damping and smoothing parameters to determine their optimal values using the L-curve method. We also solved for the event term at each grid level after identifying the most appropriate tomography parameters from the above step. At each grid level, we removed phase arrivals whose travel-time residuals exceeded 3 standard deviations from the mean. We observed 7.6% reduction in root mean square from the 1° grid to the 0.125° grid and as the grid spacings decreases, robustness of smaller tectonic features in data-rich regions such as western US, Alaska, western Europe, Middle East, and Japan vastly improves, increasing the confidence of the model for accurate event location. We are conducting Pn travel time prediction using the global RSTT models developed using the variable grid refinement method.
Session: Advancements in Forensic Seismology and Explosion Monitoring - I
Type: Oral
Date: 4/17/2025
Presentation Time: 09:00 AM (local time)
Presenting Author: Nishath
Student Presenter: No
Invited Presentation:
Poster Number:
Authors
Nishath Ranasinghe Presenting Author Corresponding Author ranasinghe@lanl.gov Los Alamos National Laboratory |
Michael Begnaud mbegnaud@lanl.gov Los Alamos National Laboratory |
Charlotte Rowe char@lanl.gov Los Alamos National Laboratory |
Stephen Myers myers30@llnl.gov Lawrence Livermore National Laboratory |
Brain Young byoung@sandia.gov Sandia National Laboratories |
|
|
|
|
Variable Global Grid Refinement and Prediction Using RSTT Model
Category
Advancements in Forensic Seismology and Explosion Monitoring