Comparing Sensitivities of Geodetic Processing Methods for Rapid Earthquake Magnitude Estimation

Abstract

Rapid earthquake magnitude estimation from real-time space-based geodetic observation streams provides an opportunity to mitigate the impact of large and potentially damaging earthquakes by issuing low-latency warnings prior to any significant and destructive shaking. Geodetic contributions to earthquake characterization and rapid magnitude estimation have evolved in the last 20 yr, from post-processed seismic waveforms to, more recently, improved capacity of regional geodetic networks enabled real-time Global Navigation Satellite System seismology using precise point positioning (PPP) displacement estimates. In addition, empirical scaling laws relating earthquake magnitude to peak ground displacement (PGD) at a given hypocentral distance have proven effective in rapid earthquake magnitude estimation, with an emphasis on performance in earthquakes larger than ∼Mw 6.5 in which near-field seismometers generally saturate. Although the primary geodetic contributions to date in earthquake early warning have focused on the use of 3D position estimates and displacements, concurrent efforts in time-differenced carrier phase (TDCP)-derived velocity estimates also have demonstrated that this methodology has utility, including similarly derived empirical scaling relationships. This study builds upon previous efforts in quantifying the ambient noise of three-component ground-displacement and ground-velocity estimates. We relate these noise thresholds to expected signals based on published scaling laws. Finally, we compare the performance of PPP-derived PGD to TDCP-derived peak ground velocity (PGV), given several rich event datasets. Our results indicate that TDCP-PGV is more likely than PPP-PGD to detect intermediate magnitude (∼Mw 5.0–6.0) earthquakes, albeit with greater magnitude estimate uncertainty and across smaller epicentral distances. We conclude that the computationally lightweight TDCP-derived PGV magnitude estimation is complementary to PPP-derived PGD magnitude estimates, which could be produced at the network edge at high rates and with increased sensitivity to ground motion than current PPP estimates.

Supervised Machine Learning of High Rate GNSS Velocities for Earthquake Strong Motion Signals

“…(2020) and Dittmann et al. (2022), we estimated the noise threshold in the 2 min window prior to seismic origin time. Hodgkinson et al.…”

“…Threshold is the NCX2‐995 approach used by Dittmann et al. (2022) that thresholds the noise based upon the 0.995 significance of a non‐central chi‐square distribution of the ambient noise. STA/LTA is based on Psimoulis et al.…”

“…We processed these 5 hz measurements with the open‐source SNIVEL package (Crowell, 2021) using broadcast ephemeris and narrow lane phase combinations. We chose TDCP over PPP because it is more sensitive to motion (Dittmann et al., 2022; Fang et al., 2020), and it is “lightweight” in that it does not require sophisticated corrections and is computationally inexpensive. From a machine learning perspective, this could be considered a first step in our feature engineering, or applying domain knowledge to extracting features that are correlated with motion in observed carrier phase measurements.…”

“…Logic was applied to map existing continuous epoch-wise time domain threshold detection to our 30 s overlapping window target labels. For a threshold method comparison similar to the approach of Hodgkinson et al (2020) and Dittmann et al (2022), we estimated the noise threshold in the 2 min window prior to seismic origin time. Hodgkinson et al (2020) characterized the stand-alone sensitivity of detection using ambient noise antecedent to an event as a Gaussian heuristic threshold.…”

“…Recent interest in peak geodetic dynamic signals (Crowell, 2021; Fang et al., 2020; Melgar et al., 2015; Ruhl et al., 2019) prompted use of unfiltered timeseries to capture peak signals for magnitude scaling laws and ground motion intensity measurements (Figures 1b and 1c). These epoch‐wise threshold detection methods (i.e., Crowell et al., 2009; Dittmann et al., 2022; Hohensinn & Geiger, 2018; Hodgkinson et al., 2020; Psimoulis et al., 2018) use instantaneous measurements to estimate motion onset and therefore are a step‐forward for inclusion of GNSS‐seismology waveforms, but have limited “real‐world” testing and most importantly mitigate false alerts for operational systems by correlating detections with proximal stations within networks or windowing in time from seismic triggers. These processes reduce the usefulness of these measurements for rapid, stand‐alone decision criteria.…”