The Great Sumatra-Andaman Islands Earthquake and Tsunami of 26 December 2004 IV
Presiding: R S Gross, Jet Propulsion Laboratory, California Institute of Technology; Y Bock, Scripps Institution of Oceanography
U52A-01 INVITED 10:30h
Effects of the Mw 9.0 Sumatra Earthquake and Tsunami on Earth's Shape
Global Positioning System (GPS) observations from the global IGS network are used to investigate geodynamic processes induced by the Mw 9.0 Sumatra earthquake and its subsequent tsunami. The earthquake would have induced change in Earth's shape over a broad spatio-temporal spectrum, including strong motion; permanent co-seismic displacement; surface loading due to the tsunami propagating barotropically through the global ocean; free oscillations of the solid Earth with periods of up to 53 minutes lasting for several days after the initial earthquake; and transient post-seismic deformations lasting from days to years. Initial model calculations of the co-seismic displacement field predict horizontal displacements of >20 mm in the near field (up to 1000 km) and >0.1 mm for the entire Earth's surface. Tsunami-induced loading signals are expected to be of the order of several mm. The free oscillations may have vertical amplitudes of several cm. GPS data are analyzed to determine the global pattern of the 3-d displacement field induced by the earthquake. GPS-determined time series of 3-d displacements with high temporal resolution will be analyzed to determine the geometric amplitudes of the free oscillations and the tsunami load-induced deformations of the earth surface. For the latter, predictions based on modeled sea-level anomalies will be fitted to the global polyhedron of observed displacements. The ultimate goal is to assess the feasibility of using the tsunami-induced changes in the Earth's shape as observed by the regional/global GPS network in improving tsunami models and having it incorporated into a tsunami warning system.
U52A-02 10:45h
Excitation of Earth Rotation and Gravitational Field Changes by the December 26, 2004 Sumatran Earthquake
Besides generating seismic waves, which eventually dissipate, an earthquake also generates a static displacement field everywhere within the Earth. This global displacement field rearranges the Earth's mass, causing the Earth's rotation and gravitational field to change. The size of these changes depends upon the size, focal mechanism, and location of the earthquake. The Sumatran earthquake of December 26, 2004 is the largest earthquake to have occurred since the 1964 Alaskan earthquake. The coseismic effect of the Sumatran earthquake upon the Earth's length-of-day, polar motion, and low-degree harmonic coefficients of the gravitational field are computed. It is found that this earthquake should have caused the length-of-day to decrease by 2.68 Îœsec, the position of the mean rotation pole to shift 0.821 mas towards 145°E longitude, the Earth's oblateness J2 to decrease by 0.90×10-11, and the Earth's pear-shapedness J3 to decrease by 0.19×10-11. This predicted change in the length-of-day is probably not detectable by current measurement systems, the change in oblateness is perhaps just detectable, and the change in the mean position of the rotation pole is perhaps detectable if other effects, such as those of the atmosphere, oceans, and continental water storage, can be adequately removed from the observations.
U52A-03 11:00h
The Effects of the 2004 Sumatra Earthquake on Earth Rotation: A Comparison Between Theory and Different Observational Techniques
On 26 December 2004 a large earthquake (Mw=9) shook northern Sumatra, Indonesia, and then shook the whole of humanity by causing probably the deadliest tsunami in recorded history. Conservation of angular momentum, a fundamental principle of physics, indicates that the earthquake also had an effect on Earth rotation. Current theory estimates that this event shifted the mean pole approximately 2.5 cm in the direction of 145E and decreased the length of day by 2.68 microseconds [Chao and Gross, EOS, 86(1), 2005]. This study compares these theoretical results to Earth orientation parameters (EOP) determined using different observational-based techniques and estimates produced by combining these various observational techniques. These various techniques include the Global Positioning System (GPS), Very Long Baseline Interferometry (VLBI), and Satellite Laser Ranging (SLR). For example, daily estimates of EOP from an improved analysis of residuals between numerically-integrated inertial GPS satellite orbits and observed Earth-referenced GPS satellite positions can serve as a source of information on EOP as they actually were after the earthquake. These observational results are then compared to EOP predictions made before the earthquake by the Rapid Service/Prediction Center of the International Earth Rotation and Reference Systems Service, located at the U.S. Naval Observatory. This propagation method can take observations of EOP made before the earthquake and extend them through December 26, giving EOP as they would have been if the earthquake had not occurred plus some amount of prediction error. The uncertainties in the methods of estimating EOP and in the prediction method are briefly discussed. On the basis of the differences between observed EOP after the earthquake and EOP predicted prior to the earthquake, conclusions are drawn concerning both the validity of the theoretical results and the observability of large earthquakes in temporal variations of Earth orientation.
U52A-04 11:15h
Analysis of Satellite Time-lapse and Sunglint Imagery of Tsunami Waves from the 26 December 2004 Great Sumatra-Andaman Islands Earthquake
The magnitude 9.0 earthquake which occurred off the west coast of Sumatra, Indonesia at 00:58:53 UTC on 26 December 2004 triggered a major tsunami, which caused catastrophic devastation and loss of life in the Indian Ocean region. The leading tsunami wave moved through the waters of the Indian Ocean at approximately 750 km per hour, reaching the eastern coast of India around 3:35 UTC, based on tide gauge measurements at the Indian port of Visakhapatnam. It is well known, however, that tsunamis occur as a series of large waves that continue to arrive some time after the initial wave, often causing significant additional destruction. Between 5:10 and 5:20 UTC the Multi-angle Imaging SpectroRadiometer (MISR) aboard NASA's Terra satellite detected wave activity near the eastern Indian and western Sri Lankan coasts. Because MISR's nine cameras observe any given region over a period of approximately 7 minutes, the instrument was able to capture unique time-lapse, animated imagery of extremely large waves breaking at various points along the Indian coast, including "edge waves" moving parallel to the shoreline. These data provide quantitative information on the location and timing of the breaking waves, along with estimates of speed (45 kph onshore), orientation, crest-to-crest wavelength (9 km), and frequency (12 minutes between waves). In addition, by looking at specific view angles strongly influenced by sunglint, MISR detects changes in sea surface slope. The latter technique resulted in the unexpected detection of deep ocean wave features about 30-40 km from Sri Lanka's southwestern coast, in a pattern highly correlated with the location of the continental shelf. In conjunction with bathymetric measurements of ocean depth, the information provided by MISR will be used to refine and calibrate tsunami propagation models. Improving these models has two primary benefits. First, a detailed understanding of wave interactions with coastal areas is necessary for understanding tsunami wave inundation and for developing damage mitigation approaches. Second, increased predictive capability will make possible more accurate near-real-time forecasts of tsunami arrival times and effects.
U52A-05 11:30h
Multi-sensor investigation of the Sumatran Tsunami: observations and analysis of hydroacoustic, seismic, infrasonic, and tide gauge data
We present an analysis of the acoustic signals from the December 26, 2004 Sumatra earthquakes, in conjunction with the seismic and tide gauge information from the event. The M9.0 mainshock and its aftershocks were recorded by a suite of seismic sensors around the globe, giving us information on its location and the source process. Recently installed sensor assets in the Indian Ocean have enabled us to study additional features of this significant event. Hydroacoustic signals were recorded by three hydrophone arrays, and the direction finding capability of these arrays allows us to examine the location, time and extent of the T-wave generation process. We detect a clear variation of the back-azimuth that is consistent with the spatial extent of the source rupture. Recordings from nearly co-located seismometers provide insights into the acoustic-to-seismic conversion process for T-waves at islands, along with the variation in signal characteristics with source size. Two separate infrasound arrays detect the atmospheric signals generated by the event, along with additional observations of the seismic surface wave and the T-phase. We will present a comparison of the signals from the mainshock, as a function of location and size, with those from aftershocks and similar events in the nearby region. Our acoustic observations compare favorably with model predictions of wave propagation in the region. For the hydroacoustic data, the azimuth, arrival time, and signal blockage characteristics, from three separate arrays, associate the onset of the signal with the mainshock and with a time extent consistent with the rupture propagation. Our analysis of the T-phase travel times suggests that the seismic-to-acoustic conversion occurs more than 100 km from the epicenter. The infrasound signal's arrival time and signal duration are consistent with both stratospheric and thermospheric propagation from a source region near the mainshock. We use the tide gauge data from stations around the Indian Ocean to identify the arrival time of the Tsunami. The acoustic and seismic signals associated with the earthquakes arrive at the remote stations significantly ahead of the Tsunami. We combine the information from the various sensors to investigate the ability of the acoustic stations to detect the Tsunami.
U52A-06 11:45h
Change in Meteorological and Atmospheric Parameters Associated with Sumatra Earthquake of December 26, 2004
Due to strong coupling between land, ocean and atmosphere, significant changes have been observed with recent coastal earthquakes. Various land, ocean, atmospheric and ionospheric parameters are found to show changes prior to the earthquake events that may be associated with the build up of stress in the epicentral region. In the present paper, we have carried out detailed analysis of multi sensor (optical and microwave remote sensing) data and have found changes in various atmospheric, meteorological and ocean parameters in the region around the Indian Ocean and the Bay of Bengal prior and after the Sumatra earthquake event of December 26, 2004. These parameters are compared with a low magnitude earthquake event. The influence of strong tsunami generated by the Sumatra earthquake show strong and characteristic behavior of atmospheric and meteorological parameters. These parameters may provide early information about strong earthquakes if the seismically active oceanic regions are monitored continuously.