Teleseismic inversion of the 2004 Sumatra-Andaman earthquake rupture process using complete Green’s functions

Abstract

Spatial and temporal variations in coseismic slip distribution are often obtained by rupture process analyses using teleseismic body waves. Many analyses using teleseismic body waves were based on the ray theory because of the very efficiently computable direct P-, S-, and major reflected waves near the source. The 2004 Sumatra-Andaman earthquake was one of the largest earthquakes recorded in history, and the data that are required for the entire rupture process analysis include later phases such as PP waves and a very long period phase called a W phase. However, calculating these later phases using the conventional ray theoretical method is difficult. Here we investigate the rupture process of the 2004 Sumatra-Andaman earthquake using complete Green's functions, including all later phases such as PP waves and W phase. We use the direct solution method, which computes complete synthetic seismograms up to 2 Hz for transversely isotropic spherically symmetric media, to calculate the Green's functions. The obtained synthetic seismograms generated fit the observed seismograms quite well from a short period to a long period. The estimated slip distribution consists of four large slip areas: the largest slip occurred in the shallow part off the Sumatra west coast with a maximum slip of approximately 29 m, the second and third largest slips occurred in the shallow and deep parts of the Nicobar region with maximum slips of approximately 8 m and 7 m, respectively, and the fourth slip occurred in the middle Andaman region with a maximum slip of approximately 6 m. The estimated average rupture velocity is 2.8 km/s, but the rupture may have slowed between the Sumatra and the shallow Nicobar slip areas, and between the Nicobar and the middle Andaman slip areas. The delayed initiation of the shallow slips in the Nicobar region may possibly have been triggered by the deeper slip in the Nicobar region. There were no distinct depth-varying properties for the shallow and deep slips in the Nicobar region, as were reported for the 2011 Tohoku-Oki and 2010 Chile earthquake.

Tsunami hazard assessment for the central and southern pacific coast of Colombia

“…The distribution of coupling coefficients supports the interpretation of Yoshimoto et al (2019), corroborating the hypothesis that the 2016 earthquake broke the 1942 source region whilst the 1979 event did not break the 1906 source region. This allows the generation of models of rupture and periods of megathrust recurrence in individual and separate source segments.…”

“…This inversion depends on the following: the relative plate velocity based on information available in Chlieh et al (2014); the coupling rate, which is obtained from each GPS station of the Colombian Geological Survey (Geo RED) network of GNSS stations by means of Green's equations; and the Euler pole for each plate, using the results of the MORVEL (Mid-Ocean Ridge Velocities) model for the Nazca and South American plates as a basis (DeMets, Gordon, and Argus 2010) block of the Andes. As a result, mid to high coupling coefficients (0.5-0.8) are observed in the source regions of the 1942, 1958, and 2016 earthquakes with a lower coupling coefficient (<0.5) in the 1979 source region (Sagiya and Mora-Páez 2019;Yoshimoto et al 2019).…”

“…For the 1906 scenario, a magnitude of Mw 8.5 was defined by inverting the tsunami waveforms and intensity data observed during the event. For the 1979 scenario, a magnitude of Mw 8.3 was defined by inverting teleseismic P-wave data to map the slip distribution using Green's functions (Yoshimoto and Yamanaka 2014), with an overlay on the interplate coupling to interpret the rupture modes and recurrence intervals of the event.…”

Slow slip events and the 2016 Te AraroaMw7.1 earthquake interaction: Northern Hikurangi subduction, New Zealand

“…We used the W-phase Geophysical Research Letters 10.1002/2017GL074776 waveform inversion technique as described in Benavente and Cummins [2013]. This method has been recently used for recovering the slip pattern for moderate to large earthquakes [Yoshimoto and Yamanaka, 2014;Benavente et al, 2016].…”

Change of stress patterns during 2004 MW 9.3 off‐Sumatra mega‐event: Insights from ridge–trench interaction for plate margin deformation

“…The mega‐thrust event had a rupture length of ~1,200 to 1,500 km, extending along the Andaman Trench, and had a maximum slip of ~15–30 m near the source region below Sumatra. Subsequent late moderate slip of 5–14 m between ~7 and 12°N latitudes near the Nicobar segment and a slow slip of 2–6 m to the north beyond 12°N were explained to be associated with tectonic segmentation of the margin (Ammon et al, ; Chlieh et al, ; Ishii et al, ; Khan & Chakraborty, ; Lay et al, ; Menke, Abend, Bach, Newman, & Levin, ; Yoshimoto & Yamanaka, ). The overall rupture history, physical properties of the descending lithosphere (Kennett & Cummins, ), back‐arc tectonics (Khan & Chakraborty, ), and moment energy release (Chlieh et al, ; Ishii et al, ) had distinctly isolated the margin between 7°N and ~12°N latitudes.…”