G44A-06
GLONASS Precise Point Positioning with Ambiguity Resolution (PPP-AR) and its Integration with GPS for Earthquake Early Warning

Thursday, 17 December 2015: 17:15
2002 (Moscone West)
Jianghui Geng, Wuhan University, Wuhan, China, Yehuda Bock, University of California San Diego, La Jolla, CA, United States, Antonio Avallone, National Institute of Geophysics and Volcanology, Rome, Italy, Athanassios Ganas, National Observatory of Athens, Athens, Greece, Diego Melgar, University of California Berkeley, Berkeley, CA, United States and Sebastian Riquelme, University of Chile, Santiago, Chile
Abstract:
GPS has been recognized as an essential technique to measure static and dynamic displacements for earthquake early warning (EEW), in particular for near-field large earthquake monitoring where broadband seismometers clip, filtering of strong-motion accelerometer data eliminates the static component, and seismic data are affected by magnitude saturation. We have developed a prototype EEW system for the U.S. West Coast using real-time precise point positioning with ambiguity resolution (PPP-AR). It includes about 200 GPS stations spanning the areas of greatest seismic risk: Cascadia, the San Francisco Bay Area and southern California, using IGS ultra-rapid orbits and a North-America based GPS network well outside the region of expected deformation to estimate satellite clock and fractional cycle biases. We have analyzed, retrospectively in real time, earthquakes from Mw5.1 to Mw9.0; we can detect dynamic displacements with a precision of about one cm, but we’ve encountered several issues that could improve our operational system. Reinitialization of the integer-cycle phase ambiguities is problematic when a (not unusual) data gap of more than about 5 minutes is encountered. Also, ambiguity resolution is less reliable when there are only a limited number (six or less) of GPS satellites available. Nowadays, GNSS is evolving to a true multi-constellation environment with GLONASS having been restored to a 24-satellite constellation in 2012. We present a method for GNSS (GPS+GLONASS) PPP-AR to improve on the performance of GPS PPP-AR in EEW. We find that GNSS-AR has higher reliability in achieving ambiguity-fixed solutions, improves vertical accuracy by 20%, reduces initialization time to less than 7 minutes from about 20 minutes, and reduces outlier rates from 1.2% to below 0.2%. We use three earthquakes to demonstrate GNSS-AR, the M7.8 2014 Iquique, Chile aftershock, and Mw 6.9 2014 North Aegean, Greece and Mw 6.3 2009 L’Aquila, Italy earthquakes.