Publications

Export 15 results:
Sort by: [ Author  (Asc)] Title Type Year
A B C D E F G H I J K L M N O P Q R S T U V W X Y Z   [Show ALL]
B
Baer, G, Sandwell D, Williams S, Bock Y, Shamir G.  1999.  Coseismic deformation associated with the November 1995, M-w=7.1 Nuweiba earthquake, Gulf of Elat (Aqaba), detected by synthetic aperture radar interferometry. Journal of Geophysical Research-Solid Earth. 104:25221-25232.   10.1029/1999jb900216   AbstractWebsite

The November 22, 1995, M-w=7.1 Nuweiba earthquake occurred along one of the left-stepping segments of the Dead Sea Transform in the Gulf of flat (Aqaba). Although it was the largest earthquake along this fault in the last few centuries, little is yet known about the geometry of the rupture, the slip distribution along it, and the nature of postseismic deformation following the main shock. In this study we examine the surface deformation pattern during the coseismic phase of the earthquake in an attempt to better elucidate the earthquake rupture process. As the entire rupture zone was beneath the waters of the Gulf, and there is very little Global Positioning System (GPS) data available in the region for the period spanning the earthquake, interferometric synthetic aperture radar (INSAR) provides the only source of information of surface deformation associated with this earthquake. We chose four synthetic aperture radar (SAR) scenes of about 90x90 km each spanning the rupture area, imaged by the ERS-1 and ERS-2 satellites. The coseismic interferograms show contours of equal satellite-to-ground range changes that correspond to surface displacements due to the earthquake rupture. Interferograms that span the earthquake by 1 week show similar fringe patterns' as those that span the earthquake by 6 months, suggesting that postseismic deformation is minor or confined to the first week after the earthquake. A high displacement gradient is seen on the western side of the Gulf, 20-40 km south of flat and Aqaba, where the total satellite-to-ground range changes are at least 15 cm. The displacement gradient is relatively uniform on the eastern side of the Gulf and the range changes are less than 10 cm. To interpret these results, we compare them to synthetic interferograms generated by elastic dislocation models with a variety of fault parameters. Although selecting the best fit fault parameters is nonunique, we are able to generate a group of simplified model interferograms that provide a reasonable fit to the coseismic interferogram and serve to constrain the location of the fault. The present analysis shows that if the rupture reached the Gulf-bottom surface, the mean sinistral slip along the fault is constrained to about 1.4 m. If surface rupture did not occur, the average sinistral slip is constrained to the range of 1.4-3 m for a fault patch buried 0-4 km below the Gulf-bottom Surface, respectively, with a minor normal component.

Barbot, S, Fialko Y, Sandwell D.  2008.  Effect of a compliant fault zone on the inferred earthquake slip distribution. Journal of Geophysical Research-Solid Earth. 113   10.1029/2007jb005256   AbstractWebsite

We present a new semi-analytic method to evaluate the deformation due to a screw dislocation in arbitrarily heterogeneous and/or anisotropic elastic half plane. The method employs integral transformations to reduce the governing partial differential equations to the integral Fredholm equation of the second kind. Dislocation sources, as well as spatial perturbations in the elastic properties are modeled using equivalent body forces. The solution to the Fredholm equation is obtained in the Fourier domain using a method of successive over-relaxation, and is mapped into the spatial domain using the inverse Fast Fourier Transform. We apply this method to investigate the effect of a soft damage zone around an earthquake fault on the co-seismic displacement field, and on the earthquake slip distribution inferred from inversions of geodetic data. In the presence of a kilometer-wide damage zone with a reduction of the effective shear modulus of a factor of 2, inversions that assume a laterally homogeneous model tend to underestimate the amount of slip in the middle of the seismogenic layer by as much as 20%. This bias may accentuate the inferred maxima in the seismic moment release at depth between 3-6 km suggested by previous studies of large strike-slip earthquakes.

F
Fialko, Y, Sandwell D, Agnew D, Simons M, Shearer P, Minster B.  2002.  Deformation on nearby faults induced by the 1999 Hector Mine earthquake. Science. 297:1858-1862.   10.1126/science.1074671   AbstractWebsite

Interferometric Synthetic Aperture Radar observations of surface deformation due to the 1999 Hector Mine earthquake reveal motion on several nearby faults of the eastern California shear zone. We document both vertical and horizontal displacements of several millimeters to several centimeters across kilometer-wide zones centered on pre-existing faults. Portions of some faults experienced retrograde (that is, opposite to their long-term geologic slip) motion during or shortly after the earthquake. The observed deformation likely represents elastic response of compliant fault zones to the permanent co-seismic stress changes. The induced fault displacements imply decreases in the effective shear modulus within the kilometer-wide fault zones, indicating that the latter are mechanically distinct from the ambient crustal rocks.

K
Kaneko, Y, Fialko Y, Sandwell DT, Tong X, Furuya M.  2013.  Interseismic deformation and creep along the central section of the North Anatolian Fault (Turkey): InSAR observations and implications for rate-and-state friction properties. Journal of Geophysical Research-Solid Earth. 118:316-331.   10.1029/2012jb009661   AbstractWebsite

We present high-resolution measurements of interseismic deformation along the central section of the North Anatolian Fault (NAF) in Turkey using interferometric synthetic aperture radar data from the Advanced Land Observing Satellite and Envisat missions. We generated maps of satellite line-of-sight velocity using five ascending Advanced Land Observing Satellite tracks and one descending Envisat track covering the NAF between 31.2 degrees E and 34.3 degrees E. The line-of-sight velocity reveals discontinuities of up to similar to 5 mm/yr across the Ismetpasa segment of the NAF, implying surface creep at a rate of similar to 9 mm/yr; this is a large fraction of the inferred slip rate of the NAF (21-25 mm/yr). The lateral extent of significant surface creep is about 75 km. We model the inferred surface velocity and shallow fault creep using numerical simulations of spontaneous earthquake sequences that incorporate laboratory-derived rate and state friction. Our results indicate that frictional behavior in the Ismetpasa segment is velocity strengthening at shallow depths and transitions to velocity weakening at a depth of 3-6 km. The inferred depth extent of shallow fault creep is 5.5-7 km, suggesting that the deeper locked portion of the partially creeping segment is characterized by a higher stressing rate, smaller events, and shorter recurrence interval. We also reproduce surface velocity in a locked segment of the NAF by fault models with velocity-weakening conditions at shallow depth. Our results imply that frictional behavior in a shallow portion of major active faults with little or no shallow creep is mostly velocity weakening. Citation: Kaneko, Y., Y. Fialko, D. T. Sandwell, X. Tong, and M. Furuya (2013), Interseismic deformation and creep along the central section of the North Anatolian Fault (Turkey): InSAR observations and implications for rate-and-state friction properties, J. Geophys. Res. Solid Earth, 118, 316-331, doi: 10.1029/2012JB009661.

L
Luttrell, K, Sandwell D.  2010.  Ocean loading effects on stress at near shore plate boundary fault systems. Journal of Geophysical Research-Solid Earth. 115   10.1029/2009jb006541   AbstractWebsite

Changes in eustatic sea level since the Last Glacial Maximum create a differential load across coastlines globally. The resulting plate bending in response to this load alters the state of stress within the lithosphere within a half flexural wavelength of the coast. We calculate the perturbation to the total stress tensor due to ocean loading in coastal regions. Our stress calculation is fully 3-D and makes use of a semianalytic model to efficiently calculate stresses within a thick elastic plate overlying a viscoelastic or fluid half-space. The 3-D stress perturbation is resolved into normal and shear stresses on plate boundary fault planes of known orientation so that Coulomb stress perturbations can be calculated. In the absence of complete paleoseismic indicators that span the time since the Last Glacial Maximum, we investigate the possibility that the seismic cycle of coastal plate boundary faults was affected by stress perturbations due to the change in sea level. Coulomb stress on onshore transform faults, such as the San Andreas and Alpine faults, is increased by up to 1-1.5 MPa, respectively, promoting failure primarily through a reduction in normal stress. These stress perturbations may perceptibly alter the seismic cycle of major plate boundary faults, but such effects are more likely to be observed on nearby secondary faults with a lower tectonic stress accumulation rate. In the specific instance of rapid sea level rise at the Black Sea, the seismic cycle of the nearby North Anatolian fault was likely significantly advanced.

Luttrell, K, Sandwell D, Smith-Konter B, Bills B, Bock Y.  2007.  Modulation of the earthquake cycle at the southern San Andreas fault by lake loading. Journal of Geophysical Research-Solid Earth. 112   10.1029/2006jb004752   AbstractWebsite

Changes in the level of ancient Lake Cahuilla over the last 1500 years in the Salton Trough alter the state of stress by bending the lithosphere in response to the applied lake load and by varying the pore pressure magnitude within the crust. The recurrence interval of the lake is similar to the recurrence interval of rupture on the southern San Andreas and San Jacinto faults, both of which are partially covered by the lake at its highstand. Furthermore, four of the last five ruptures on the southern San Andreas fault have occurred near a time of substantial lake level change. We investigate the effect of Coulomb stress perturbations on local faults due to changing level of Lake Cahuilla to determine a possible role for the lake in affecting the timing of fault rupture. Coulomb stress is calculated with a three-dimensional model of an elastic plate overlying a viscoelastic half-space. Plate thickness and half-space relaxation time are adjusted to match observed vertical deformation since the last lake highstand. The lake cycle causes positive and negative Coulomb stress perturbations of 0.2-0.6 MPa on the southern San Andreas within the lake and 0.1-0.2 MPa on the southern San Andreas outside the lake. These Coulomb stress perturbations are comparable to stress magnitudes known to have triggered events at other faults along the North America-Pacific plate boundary.

M
Mellors, RJ, Sichoix L, Sandwell DT.  2002.  Lack of precursory slip to the 1999 Hector Mine, California, earthquake as constrained by InSAR. Bulletin of the Seismological Society of America. 92:1443-1449.   10.1785/0120010244   AbstractWebsite

We looked for evidence of interseismic strain occurring between the 1992 Landers earthquake and the 1999 Hector Mine earthquake near the Lavic Lake and Bullion faults by using interferometric synthetic aperture radar (InSAR). Interferograms covering the Hector Mine epicentral region were studied for possible slip along the Bullion and Lavic Lake faults by both visual inspection and a matched filter technique intended to emphasize slip located at the nucleation point. Some indications of possible deformation associated with the 5 July 1992 M-L 5.4 Pisgah event was observed, but high decorrelation prevented a conclusive determination. We have seen no evidence for precursory slip in the epicentral region up to 30 days before the Hector Mine event. We estimated that the slip equivalent to a M-w 4.5 event would have been observable in the months before the Hector Mine event, and this places an upper bound on the long-term precursory slip, had it occurred. We have noted that InSAR is well suited for detecting precursory slip in general due to the high spatial resolution and the lack of ground instrumentation required but that the detection level depends on the depth and orientation of the slip.

P
Price, EJ, Sandwell DT.  1998.  Small-scale deformations associated with the 1992 Landers, California, earthquake mapped by synthetic aperture radar interferometry phase gradients. Journal of Geophysical Research-Solid Earth. 103:27001-27016.   10.1029/98jb01821   AbstractWebsite

The Landers earthquake (M-w 7.3) occurred on June 28, 1992, and ruptured nearly 100 km of previously mapped and unmapped faults in the Mojave Desert. We use synthetic aperture radar interferometry (InSAR) to examine the cumulative surface deformation between April 24 and August 7, 1992, in a 100 x 100 km region surrounding the northern portion of the earthquake rupture. Also, we introduce a technique for manipulating SAR interferograms to extract short-wavelength displacement information. This technique involves computation and subsequent combination of interferometric phase gradient maps. The InSAR results show significant deformation signatures associated with faults, fractures, dry lake beds, and mountainous regions within 75-100 km of the main rupture. Using the phase gradient method, we are able to extract small-scale deformation patterns near the main rupture. Many of the preexisting, mapped faults within 50 km of the main rupture experienced triggered slip; these include the Old Woman, Lenwood, Johnson Valley, West Calico, and Calico Faults. The InSAR results also indicate right-lateral offsets along secondary fractures trending N-NE within the left-lateral zone of shear between the main rupture and the Johnson Valley Fault. Additionally, there are interesting interferogram fringe signatures surrounding Troy Dry Lake and Coyote Dry Lake that are related to deformation of dry lake beds.

S
Sandwell, D, Smith-Konter B.  2018.  Maxwell: A semi-analytic 4D code for earthquake cycle modeling of transform fault systems. Computers & Geosciences. 114:84-97.   10.1016/j.cageo.2018.01.009   AbstractWebsite

We have developed a semi-analytic approach (and computational code) for rapidly calculating 3D time-dependent deformation and stress caused by screw dislocations imbedded within an elastic layer overlying a Maxwell viscoelastic half-space. The maxwell model is developed in the Fourier domain to exploit the computational advantages of the convolution theorem, hence substantially reducing the computational burden associated with an arbitrarily complex distribution of force couples necessary for fault modeling. The new aspect of this development is the ability to model lateral variations in shear modulus. Ten benchmark examples are provided for testing and verification of the algorithms and code. One final example simulates interseismic deformation along the San Andreas Fault System where lateral variations in shear modulus are included to simulate lateral variations in lithospheric structure.

Sandwell, DT, Price EJ.  1998.  Phase gradient approach to stacking interferograms. Journal of Geophysical Research-Solid Earth. 103:30183-30204.   10.1029/1998jb900008   AbstractWebsite

The phase gradient approach is used to construct averages and differences of interferograms without phase unwrapping. Our objectives for change detection are to increase fringe clarity and decrease errors due to tropospheric and ionospheric delay by averaging many interferograms. The standard approach requires phase unwrapping, scaling the phase according to the ratio of the perpendicular baseline, and finally forming the average or difference; however, unique phase unwrapping is usually not possible. Since the phase gradient due to topography is proportional to the perpendicular baseline, phase unwrapping is unnecessary prior to averaging or differencing. Phase unwrapping may be needed to interpret the results, but it is delayed until all of the largest topographic signals are removed. We demonstrate the method by averaging and differencing six interferograms having a suite of perpendicular baselines ranging from 18 to 406 m. Cross-spectral analysis of the difference between two Tandem interferograms provides estimates of spatial resolution, which are used to design prestack filters. A wide range of perpendicular baselines provides the best topographic recovery in terms of accuracy and coverage. Outside of mountainous areas the topography has a relative accuracy of better than 2 m. Residual interferograms (single interferogram minus stack) have tilts across the unwrapped phase that are typically 50 mm in both range and azimuth, reflecting both orbit error and atmospheric delay. Smaller-scale waves with amplitudes of 15 mm are interpreted as atmospheric lee waves. A few Global Positioning System (GPS) control points within a Game could increase the precision to similar to 20 mm for a single interferogram; further improvements may be achieved by stacking residual interferograms.

Smith, B, Sandwell D.  2004.  A three-dimensional semianalytic viscoelastic model for time-dependent analyses of the earthquake cycle. Journal of Geophysical Research-Solid Earth. 109   10.1029/2004jb003185   AbstractWebsite

[ 1] Exploring the earthquake cycle for large, complex tectonic boundaries that deform over thousands of years requires the development of sophisticated and efficient models. In this paper we introduce a semianalytic three-dimensional (3-D) linear viscoelastic Maxwell model that is developed in the Fourier domain to exploit the computational advantages of the convolution theorem. A new aspect of this model is an analytic solution for the surface loading of an elastic plate overlying a viscoelastic half-space. When fully implemented, the model simulates ( 1) interseismic stress accumulation on the upper locked portion of faults, ( 2) repeated earthquakes on prescribed fault segments, and ( 3) the viscoelastic response of the asthenosphere beneath the plate following episodic ruptures. We verify both the analytic solution and computer code through a variety of 2-D and 3-D tests and examples. On the basis of the methodology presented here, it is now possible to explore thousands of years of the earthquake cycle along geometrically complex 3-D fault systems.

T
Tong, X, Sandwell DT, Schmidt DA.  2018.  Surface creep rate and moment accumulation rate along the Aceh segment of the Sumatran Fault from L-band ALOS-1/PALSAR-1 observations. Geophysical Research Letters. 45:3404-3412.   10.1002/2017gl076723   AbstractWebsite

We analyzed the interferometric synthetic aperture radar data from the ALOS-1/PALSAR-1 satellite to image the interseismic deformation along the Sumatran fault. The interferometric synthetic aperture radar time series analysis reveals up to similar to 20 mm/year of aseismic creep on the Aceh segment along the Northern Sumatran fault. This is a large fraction of the total slip rate across this fault. The spatial extent of the aseismic creep extends for similar to 100 km. The along-strike variation of the aseismic creep has an inverse "U" shape. An analysis of the moment accumulation rate shows that the central part of the creeping section accumulates moment at approximately 50% of the rate of the surrounding locked segments. An initial analysis of temporal variations suggests that the creep rate may be decelerating with time, suggesting that the creep rate is adjusting to a stress perturbation from nearby seismic activity. Our study has implications to the earthquake hazard along the northern Sumatran fault.

Tong, XP, Sandwell D, Luttrell K, Brooks B, Bevis M, Shimada M, Foster J, Smalley R, Parra H, Soto JCB, Blanco M, Kendrick E, Genrich J, Caccamise DJ.  2010.  The 2010 Maule, Chile earthquake: Downdip rupture limit revealed by space geodesy. Geophysical Research Letters. 37   10.1029/2010gl045805   AbstractWebsite

Radar interferometry from the ALOS satellite captured the coseismic ground deformation associated with the 2010 Mw 8.8 Maule, Chile earthquake. The ALOS interferograms reveal a sharp transition in fringe pattern at similar to 150 km from the trench axis that is diagnostic of the downdip rupture limit of the Maule earthquake. An elastic dislocation model based on ascending and descending ALOS interferograms and 13 near-field 3-component GPS measurements reveals that the coseismic slip decreases more or less linearly from a maximum of 17 m (along-strike average of 6.5 m) at 18 km depth to near zero at 43-48 km depth, quantitatively indicating the downdip limit of the seismogenic zone. The depth at which slip drops to near zero appears to be at the intersection of the subducting plate with the continental Moho. Our model also suggests that the depth where coseismic slip vanishes is nearly uniform along the strike direction for a rupture length of similar to 600 km. The average coseismic slip vector and the interseismic velocity vector are not parallel, which can be interpreted as a deficit in strike-slip moment release. Citation: Tong, X., et al. (2010), The 2010 Maule, Chile earthquake: Downdip rupture limit revealed by space geodesy, Geophys. Res. Lett., 37, L24311, doi:10.1029/2010GL045805.

Tong, XP, Sandwell DT, Fialko Y.  2010.  Coseismic slip model of the 2008 Wenchuan earthquake derived from joint inversion of interferometric synthetic aperture radar, GPS, and field data. Journal of Geophysical Research-Solid Earth. 115   10.1029/2009jb006625   AbstractWebsite

We derived a coseismic slip model for the M(w) 7.9 2008 Wenchuan earthquake on the basis of radar line-of-sight displacements from ALOS interferograms, GPS vectors, and geological field data. Available interferometric synthetic aperture radar (InSAR) data provided a nearly complete coverage of the surface deformation along both ascending (fine beam mode) and descending orbits (ScanSAR to ScanSAR mode). The earthquake was modeled using four subfaults with variable geometry and dip to capture the simultaneous rupture of both the Beichuan fault and the Pengguan fault. Our model misfits show that the InSAR and GPS data are highly compatible; the combined inversion yields a 93% variance reduction. The best fit model has fault planes that rotate from shallow dip in the south (35 degrees) to nearly vertical dip toward the north (70 degrees). Our rupture model is complex with variations in both depth and rake along two major fault strands. In the southern segment of the Beichuan fault, the slip is mostly thrust (<13 m) and occurred principally in the upper 10 km of the crust; the rupture progressively transformed to right-lateral strike slip as it propagated northeast (with maximum offsets of 7 m). Our model suggests that most of the moment release was limited to the shallow part of the crust (depth less than 10 km). We did not find any "shallow slip deficit" in the slip depth distribution of this mixed mechanism earthquake. Aftershocks were primarily distributed below the section of the fault that ruptured coseismically.

X
Xu, XH, Ward LA, Jiang JL, Smith-Konter B, Tymofyeyeva E, Lindsey EO, Sylvester AG, Sandwell DT.  2018.  Surface creep rate of the southern San Andreas Fault modulated by stress perturbations from nearby large events. Geophysical Research Letters. 45:10259-10268.   10.1029/2018gl080137   AbstractWebsite

A major challenge for understanding the physics of shallow fault creep has been to observe and model the long-term effect of stress changes on creep rate. Here we investigate the surface creep along the southern San Andreas fault (SSAF) using data from interferometric synthetic aperture radar spanning over 25 years (ERS 1992-1999, ENVISAT 2003-2010, and Sentinel-1 2014-present). The main result of this analysis is that the average surface creep rate increased after the Landers event and then decreased by a factor of 2-7 over the past few decades. We consider quasi-static and dynamic Coulomb stress changes on the SSAF due to these three major events. From our analysis, the elevated creep rates after the Landers can only be explained by static stress changes, indicating that even in the presence of dynamically triggered creep, static stress changes may have a long-lasting effect on SSAF creep rates. Plain Language Summary There are two significant conclusions from this study. First, we analyzed 25 years of InSAR measurements over the Southern San Andreas Fault system to document a major increase in the average creep rate following the 1992 Mw 7.3 Landers Earthquake which is then followed by creep rate reductions after the 1999 Mw 7.1 Hector Mine Earthquake and the 2010 Mw 7.2 El Major Cucapah Earthquake. Second, we attribute all these creep rate changes to the Coulomb stress variations from these three major Earthquakes. The dynamic Coulomb stress changes are similar for all three events, contributing to triggered creep on the SSAF. In contrast, the static Coulomb stress changes on the SSAF are positive after the Landers and negative after the Hector Mine and El Major Cucapah, coinciding with the higher average creep rate after the Landers and lower rates after the other two events. An implication of this study is that small but steady Coulomb stress changes have a larger impact on shallow creep than the larger dynamic stress changes associated with passing seismic waves. These results illuminate the significance of time scale-dependent complexity of shallow fault creep and how these behaviors are communicated by stress perturbations from regional earthquakes.