Moment magnitudes for large earthquakes (Mw≥7.0) derived in real time from near-field seismic data can be underestimated due to instrument limitations, ground tilting, and saturation of frequency/amplitude-magnitude relationships. Real-time high-rate GPS resolves the buildup of static surface displacements with the S wave arrival (assuming nonsupershear rupture), thus enabling the estimation of slip on a finite fault and the event's geodetic moment. Recently, a range of high-rate GPS strategies have been demonstrated on off-line data. Here we present the first operational system for real-time GPS-enhanced earthquake early warning as implemented at the Berkeley Seismological Laboratory (BSL) and currently analyzing real-time data for Northern California. The BSL generates real-time position estimates operationally using data from 62 GPS stations in Northern California. A fully triangulated network defines 170+ station pairs processed with the software trackRT. The BSL uses G-larmS, the Geodetic Alarm System, to analyze these positioning time series and determine static offsets and preevent quality parameters. G-larmS derives and broadcasts finite fault and magnitude information through least-squares inversion of the static offsets for slip based on a priori fault orientation and location information. This system tightly integrates seismic alarm systems (CISN-ShakeAlert, ElarmS-2) as it uses their P wave detections to trigger its processing; quality control runs continuously. We use a synthetic Hayward Fault earthquake scenario on real-time streams to demonstrate recovery of slip and magnitude. Reanalysis of the Mw7.2 El Mayor-Cucapah earthquake tests the impact of dynamic motions on offset estimation. Using these test cases, we explore sensitivities to disturbances of a priori constraints (origin time, location, and fault strike/dip). link to publication