Rupture Process of the 1999 Mw 7.1 Duzce Earthquake from Joint Analysis of SPOT, GPS, InSAR, Strong-Motion, and Teleseismic Data: A Supershear Rupture with Variable Rupture Velocity

Abstract

We analyze the rupture process of the 1999 M w 7.1 Duzce earthquake using seismological, remote sensing, and geodetic data. Ground deformation measured from the subpixel cross correlation of Satellite Pour l'Observation de la Terre (SPOT) images reveals a 55 km long fault trace and smooth surface-slip distribution peaking at 3.5-4 m. The westernmost segment overlaps for over 10 km with ruptures from the M w 7.4 Izmit earthquake. The 15 km long easternmost segment, which cuts across mountainous topography, had not been reported previously. We determine a well-constrained source model using a four-segment fault geometry using constraints on surface fault slip and inverting Global Positioning System and Interferometric Synthetic Aperture Radar data along with strong-motion records. Our results show that some variability of the rupture velocity and an eastward supershear velocity are required to fit the strong-motion data. The rise time, up to 6 sec, correlates with cumulative slip, suggesting a sliding velocity of about 1 m=sec. The source model predicts teleseismic waveforms well, although early by 2 sec. This time shift is probably due to the weak beginning of the earthquake that is not observable at teleseismic distances. Strong-motion records are relatively well predicted from a source model derived from the teleseismic data using the fault geometry derived from the satellite images. This study demonstrates the benefit of using accurate fault geometries to determine finite-fault source models.Online Material: Tables of best-fitting dip angles and rupture velocities, and figures showing mutual consistency of derived displacements fields, additional slip and rise-time models, and teleseismic waveform fit.

Seismogenic faulting of the sedimentary sequence and laterally variable material properties in the Zagros Mountains (Iran) revealed by the August 2014 Murmuri (E. Dehloran) earthquake sequence

“…We then invert for the distribution of slip on this plane that provides the closest match to the InSAR results. We discretize the fault plane into 2 km × 2 km square patches, and the inversions are performed using a simulated annealing algorithm to find the model that best fits the geodetic data (for a more complete description of the method used, see Ji et al (2002) and Konca et al (2008Konca et al ( , 2010). For computational efficiency, the InSAR data was down-sampled using a Quadtree algorithm (see Jonsson et al (2002) and Wright et al (2004) for detailed descriptions of this method).…”

“…In order to combat the ambiguity in rake that can occur when observations from only one InSAR look-angle are available, we impose the seismologically derived limits on the range of rakes that can be present in our inversions. We invert for the distribution of slip and rake that best fit the InSAR displacements using the simulated-annealing-based method described in detail by Ji et al (2002) and Konca et al (2008Konca et al ( , 2010. We do not include seismic waveforms in this inversion (as is possible using this routine) because the small spatial size of the ruptured patch means that limited additional information regarding the spatial location of rupture can be obtained by using teleseismic waveforms, and no locally recorded seismograms are available.…”

Kinematic Inversion of Physically Plausible Earthquake Source Models Obtained from Dynamic Rupture Simulations

“…This commonly used kinematic inversion code allows for the joint inversion of seismological records, both teleseismic and near field, as well as static displacements measured from both GPS and remote sensing (e.g., Ji et al, 2003Ji et al, , 2004Konca et al, 2008Konca et al, , 2010Sladen et al, 2010;. This method does not differ fundamentally from the other methods used for the joint inversion of seismological and geodetic data (e.g., Delouis et al, 2002;Liu and Archuleta, 2004).…”

“…It is possible, based on crack theory, to relate rupture velocity to fracture energy (Husseini et al, 1975;Andrews, 1976;Husseini and Randall, 1976), and hence kinematic source models can be used to estimate fracture energy and radiation efficiency of seismic ruptures. Kinematic models can also be used to demonstrate examples of supershear seismic ruptures (e.g., Bouchon et al, 2001;Konca et al, 2010), and to assess fault frictional properties (Ide and Takeo, 1997). However, this can be challenging.…”

Coseismic displacements of the 14 November 2016 <i>M</i><sub>w</sub> 7.8 Kaikoura, New Zealand, earthquake using the Planet optical cubesat constellation

“…Coseismic displacements have, for instance, been measured on repeat data from Landsat (Liu et al, 2006;Avouac et al, 2014;Barnhart et al, 2014), ASTER , SPOT (Dominguez et al, 2003;Leprince et al, 2007;Konca et al, 2010), very high-resolution optical satellites (Barnhart et al, 2015;Zhou et al, 2015), or air photos (Michel and Avouac, 2006;Ayoub et al, 2009). Coseismic displacements from Sentinel-2 data have to our best knowledge not yet been published in peer-reviewed journal publications, but are used by operational services (COMET, 2016).…”