The 26 December 2004 Sumatra earthquake (Mw 9.2-9.3) generated the most deadly tsunami in history. Yet within the first hour, the true danger of a major ocean wide tsunami was not indicated by seismic magnitude estimates, which were far too low (Mw 8.0-8.5). This problem relates to the inherent saturation of seismic wave methods designed to estimate magnitude with minutes of the event. We have shown that the earthquake's true size and tsunami potential could have been determined using Global Positioning System (GPS) data up to only 15 min after earthquake initiation, by tracking the mean displacement of the Earth's surface associated with the arrival of seismic waves. Within minutes, displacements of >10 mm are detectable as far away as India, consistent with results using weeks of data after the event. These displacements imply Mw 9.0 ± 0.1, indicating a high tsunami potential. This suggests existing GPS infrastructure could be developed into an effective component of tsunami warning systems. We present our recent findings on the design specifications a real-time GPS component of future tsunami warning systems.
It is now being proposed jointly by NASA, NOAA and USGS to exploit the increasingly available global and regional real-time GPS data from NASA's operational Global Differential GPS (GDGPS) System to enable more accurate and timely assessment of the magnitude and mechanism of large earthquakes, as well as the magnitude and direction of resulting tsunamis. The idea is to use GPS-based information, in addition to existing data types, to enhance the USGS operational system for post-earthquake damage assessment and emergency response, and to improve tsunami warnings by NOAA's Pacific Tsunami Warning Center (PTWC).
In addition, we note that the IGS Real-Time Pilot Project is just underway, which will provide real-time access to precise satellite orbits and clocks to enable rapid precise point positioning. As current geodetic techniques approach the ability to monitor ground movements with millimeter accuracy over the broadband range of ~1 second to ~10 years, it becomes apparent that IGS, and more generally the Global Geodetic Observing System (GGOS), should be exploited for geohazard prediction and early warning systems. Broadly speaking, successful prediction and early warning require two very different system designs. On the one hand, prediction systems are characterized by high accuracy measurements, detailed modeling and understanding, and long term stability to provide a standard frame of reference as a basis for prediction. On the other hand, early warning systems are characterized by real-time sensitivity and automatic response to events, and robustness against false alarms.
Since geohazards are often associated with long-term cumulative processes leading to precipitously damaging events, there is an obvious advantage if the two systems being used for prediction and early warning are developed within a self-consistent framework, as could be provided by GGOS. This way, the early warning system design can be better informed by the understanding gained from the prediction system. Prediction also helps to target the warning systems more efficiently. Precise positioning using GPS/GNSS can be done at high rate in real-time, and so can bridge the bandwidth from seconds to decades, enabling an early warning capability, while providing a connection to the more long-term stable components of GGOS required for prediction.
We suggest that the EarthScope PBO/PANGA array in the Cascadia subduction zone provides an ideal candidate location to begin implementing the concept of combined prediction/early warning systems. Our simulations (Figure 1) suggest that an Mw 9.0 earthquake in this region would generate >10 cm static displacements over a broad area covered by PBO/PANGA, which would be easily detectable within minutes of the earthquake origin time. We would therefore recommend to upgrade the PBO network to provide data in real time.