A Combined PGD–PGV Scaling Law with Rproxy Distance for the G-FAST Earthquake Early Warning Module

Early estimation of strong ground motions and earthquake magnitude is essential for earthquake and tsunami early warning systems and rapid response. We present a new magnitude scaling law relating peak ground velocities (PGVs) and peak ground displacements (PGDs) derived from Global Navigation Satellite Systems (GNSS) data based on a global dataset of real earthquake observations. We used high-rate (1-Hz) GNSS observations for earthquakes ranging in magnitude from Mw 4.9 to 9.1. With this dataset, we explored the evolution of PGV and formally combined it with the PGD scaling law that forms the core of the geodetic first approximation of size and time (G-FAST) module in ShakeAlert, by computing a new distance metric based directly on the PGV and ground-motion model equations developed from a very-high-rate GNSS dataset. We assess the validity of this new distance metric and its performance within the scaling law by comparing it with the performance of the scaling law when using other distance metrics, such as hypocentral distance and mean rupture distance. Moreover, we investigate four case study events with unique rupture characteristics and explore the potential to determine earthquake magnitude from PGV and PGD observations. We find that our newly developed distance metric characterizes true ground motion better than traditionally used distance metrics, including hypocentral distance, which is currently used in early warning operations. We finish by discussing the downstream impacts of improved ground-motion characterization in the context of operational early warning.