G43B-01
Fault geometry and slip distribution of the 10 December 2003 Taiwan Mw 6.5 earthquake from continuous GPS observations
Taiwan is located at the plate boundary zone between the Eurasian and the Philippine Sea plates where seismicity is extremely high. The MW 6.5 Chengkung earthquake occurred in eastern Taiwan at 04:38 UTC on 10 December 2003. In this paper, the GPS data collected from 18 continuously recording stations and 86 campaign stations are used to model the coseismic and postseismic deformation with the fault model in the elastic dislocation theory. It has been found that three-fault model can well show the reverse and lateral slip distribution. The inferred slip distribution indicates high slip patches near the epicenter. In addition, six months of postseismic time-series are better fit with a logarithmic instead of exponential function, suggesting that the postseismic deformation is possibly controlled by afterslip. Our inversion model predicts afterslip to be concentrated both up- and down-dip from patches of maximum coseismic slip where aftershocks are sparse. The afterslip distribution on the coseismic rupture plane is strongly dominated by reverse slip of the main center of coseismic reverse slip.
G43B-02
Semi-analytic Models of Deformation in Heterogeneous Media With Nonlinear Visco-elastic Rheologies.
Elastic properties of rocks may vary substantially (generally, in 3-D) in the seismogenic zone, and rheologic behavior of rocks below the brittle-ductile transition may be highly non-linear. Such variations present a considerable challenge for the accurate modeling of the response of the lithosphere to the co-seismic excitation. We develop a methodology to compute the time-dependent response of the lithosphere after the occurrence of an earthquake in the case of an arbitrarily heterogeneous and anisotropic upper crust underlain by a visco-elastic substrate obeying the power-law rheology. The time evolution of the deformation is approximated by a finite difference scheme involving purely elastic solutions for equivalent body forces that account for variations in the effective viscosity. It follows, for instance, that Green's functions such as the Okada's solution for a homogeneous half-space can be used to model the time-dependent deformation of a visco-elastic body with an arbitrarily varying nonlinear viscosity structure. We treat the case of an anti-plane screw dislocation for which we derive a new iterative spectral scheme based on Fredholm integrals. The static two-dimensional numerical solution is benchmarked with several analytical or semi-analytical formulations in the cases of full relaxation of a horizontal viscous substrate, and anisotropic or elastic heterogeneities across the fault. The time dependent deformation in the case of linear visco-elasticity favorably compares to the propagator matrix semi-analytical solution. We apply the Fredholm solver to compute the surface deformation following the healing of a damage zone in the vicinity of an earthquake rupture and assess a potential bias in slip inversions that results from the neglect of a fault damage zone. Finally, we show that the same methodology can be applied in three dimensions.
G43B-03
ERP-Variations on Time Scales Between Hours and Months Derived From GNSS Observations
Current observations gained by the space geodetic techniques, especially VLBI, GPS and SLR, allow for the determination of Earth Rotation Parameters (ERPs - polar motion, UT1/LOD) with unprecedented accuracy and temporal resolution. This presentation focuses on contributions to the ERP recovery provided by satellite navigation systems (primarily GPS). The IGS (International GNSS Service), for example, currently provides daily polar motion with an accuracy of less than 0.1mas and LOD estimates with an accuracy of a few microseconds. To study more rapid variations in polar motion and LOD we established in a first step a high resolution (hourly resolution) ERP-time series from GPS observation data of the IGS network covering the year 2005. The calculations were carried out by means of the Bernese GPS Software V5.0 considering observations from a subset of 113 fairly stable stations out of the IGS05 reference frame sites. From these ERP time series the amplitudes of the major diurnal and semidiurnal variations caused by ocean tides are estimated. After correcting the series for ocean tides the remaining geodetic observed excitation is compared with variations of atmospheric excitation (AAM). To study the sensitivity of the estimates with respect to the applied mapping function we applied both the widely used NMF (Niell Mapping Function) and the VMF1 (Vienna Mapping Function 1). In addition, based on computations covering two months in 2005, the potential improvement due to the use of additional GLONASS data will be discussed.
G43B-04
Impact of Solar activity, Volcanic and Greenhouse gases forcings in Atmospheric Angular Momentum variability and length of the day for the 20th century as seen in IPCC coupled simulations
Different scenarios taking into account key forcings (Solar activity, volcanic activity, different greenhouse gases, as well as combinations of these) has been selected for the IPCC for understanding their respective impact in the global climate of the 20th century and possible effect in futures climates. In this research we show the results related with the atmospheric angular momentum computed from the different model runs (more than 50 different runs) under different scenarios performed by NCAR coupled models, which simulated the 20th century climate variability. We investigated particularly the difference in the role played at decadal time scales and trend of the solar and volcanic forcings.
G43B-05
The Temporal Variation Analysis Of Cananéia GPS Station Coordinates
The objective of this paper is present the results obtained in the data processing of NEIA GPS permanent station, which belongs to the Brazilian Network of Continuous Monitoring of GPS System (RBMC), and to analyze the variations suffered of a time series of the station coordinates. The data had been collected in a double frequency GPS geodesic receiver, with 15 seconds rate of collection. The period analyzed was from January 2002 to April 2006. The data was processed with Bernese V.5.0 software and the AUSPOS on-line service, using differential method. In the Bernese processing was used the fiducial station PARA (Curitiba/Brazil) and for the AUSPOS processing was used BRAZ (Brasília - Brazil), LPGS (La Plata/Argentina) and UNSA (Salta/Argentina), these stations are in the same lithospherical plate. In the both processing's was used a propagation model of planimetric speed to detect NEIA station movement. After the processing's, it were made the difference between the calculated and the fiducial coordinates of NEIA station in the latitude, longitude and geodetic height. The displacement analysis at NEIA station, show variations of its coordinates caused by the South American Plate movement. The Bernese processing results show a displacement around 1,25 cm/year in the North direction and 0,30 cm/year in the West direction. The planimetric resultant vector displacement for NEIA station is about 1,30 cm/year in the Northwest direction. On the other hand, the AUSPOS processing service presented approximately 1,03 cm/year displacement in the North direction for latitude and for longitude an inverse displacements can be verified for the two sessions (first session: 0 to 12 hour and the second session 12 to 24 hour), 0,10 cm/year in West direction and -0,10 cm/year in East direction, respectively. Although, for the geometric height the Bernese and AUSPOS processing results, presents different signals. In the first session the results from Bernese processing present a -0,46 cm/year trend while AUSPOS service present a -0,03 cm/year trend and, for the second session, the Bernese presents -0,64 cm/year trend and the AUSPOS was about 0,03 cm/year trend. A more detailed analysis of the vertical component can be obtained if a comparison of GPS derived geodetic height and the sea level height from tide gauge station. This comparison will be fundamental for the knowledge of the variation of the sea level in the region of study.
G43B-06
Contemporary Deformation within the Snake River Plain and Northern Basin and Range Province, USA
GPS velocities, earthquakes, faults, and volcanic features are used to evaluate contemporary deformation within the Snake River Plain (SRP) and surrounding northern Basin and Range Province. The SRP is a prominent low- relief physiographic feature that extends from eastern Oregon through southern Idaho and into northwestern Wyoming, USA. The Eastern Snake River Plain (ESRP) is a 400-km long, NE-trending volcanic province that is characterized by bimodal volcanism, which represents the track of the Yellowstone Hotspot currently located in Wyoming. The Western Snake River Plain (WSRP) is a 300-km long, NW-trending graben that extends into eastern Oregon. The WSRP is an extensional basin that formed adjacent to an earlier position of the Yellowstone Hotspot in southern Idaho. Previous geodetic investigations suggest the ESRP and, perhaps the WSRP, have GPS velocities indicative of rigid block motion of the SRP along its physiographic boundaries. GPS data compiled for this study are used to test this hypothesis. Several institutions including the National Geodetic Survey, Idaho National Laboratory, Rensselaer Polytechnic Institute, and University of Utah observed GPS stations from 1994 to 2006 within the SRP and surrounding region. Horizontal velocities show generally consistent N110°W orientations with an average rate of 1.5 ± 0.3 mm/yr (for 11 stations) along most of the ESRP and adjacent northwest Basin and Range, although some Basin and Range velocities are less and may be influenced by post viscoelastic relaxation following the 1983 Mw 6.9 normal-faulting Borah Peak, Idaho earthquake. GPS velocities with an average rate of 1.9 ± 0.3 mm/yr (for 5 stations) change orientation to N95°W at a distance of 190 km from the Yellowstone Hotspot within the southern region of the ESRP and adjacent Basin and Range. Within the WSRP, GPS velocities have an average rate of 2.0 ± 0.5 mm/yr (for 7 stations) and change orientation to N40°W. These GPS velocities are more consistent with those in eastern Oregon, a region that is rotating clockwise relative to North America. To assess possible rotations and strain rates, we invert GPS horizontal velocities, geologic fault slip rates, earthquake-derived fault slip vector azimuths, and volcanic dike extension rates. We interpret GPS velocities to describe the relative motions of coherent regions of consistent strain within the SRP and surrounding Basin and Range Province.
G43B-07
Block Modeling of Crustal Deformation in the Northern Walker Lane, Western Basin and Range, to Improve Estimates of Seismic Hazard
In the United States, seismic hazard is evaluated officially by the U.S. Geological Survey and published as estimates in the National Seismic Hazard Maps (NSHM) that depict the peak ground shaking at a specific level of likelihood. In the western Great Basin, the 2002 NSHM is based on a combination of seismic, geologic and geodetic data. However, a discrepancy between the deformation rate that is inferred from the geodetic data (e.g. GPS) and geologic data (e.g. slip rates from fault studies) led to the introduction of an ad hoc zone of crustal shear strain in the western Basin and Range. Only then was the shaking risk portrayed in the NSHM consistent with the relative geodetic velocity of the Sierra Nevada microplate with respect to the central Great Basin. Since creation of the 2002 NSHM there has been a rapid increase in the quantity, quality and spatial coverage of GPS data in the western Great Basin, providing a vast improvement on the constraint on the pattern of crustal deformation. Thus geodesy is poised to make a substantial contribution to the spatial localization of seismic hazard in support of the next generation NSHM. In the Walker Lane ~10 mm/yr of relative motion are accommodated as shear and extension along a ~200 km wide and ~1000 km long zone of intracontinental deformation associated with the Pacific/North American plate boundary. We integrate GPS velocities obtained from sites in the continuous BARGEN, PBO, BARD, semi-continuous MAGNET network plus campaign results from numerous published results to constrain block models of crustal deformation. In so doing we estimate slip rates on block-bounding faults that have regional kinematic self-consistency and can be easily incorporated into the USGS algorithms that compute estimates for seismic hazard. Because of the large number and high density of candidate faults, and length of this zone we divide the region into three parts covering the Northern, Central and Southern Walker Lane. We have completed building the models for the northern Walker Lane (latitude 38.5 to 40.5 degrees, longitude - 120 to -117 degrees) and will present results from this section. The model has 62 blocks with mean dimension of ~30 km on a side, and thus the blocks are small compared to the width of the expected signal owing to elastic strain accumulation across locked faults. This starting number of blocks is purposefully large. We will discuss the ability that the data have to resolve details in the pattern of crustal deformation, and make a special effort to quantify the uncertainties and trade-offs in slip rates of nearby faults whose strain signals can overlap.
G43B-08
Space geodetic investigation of the co- and post-seismic deformation due to the 2003 Mw 7.3 Altai earthquake: Implications for the local lithospheric rheology
We use ENVISAT Advanced Synthetic Aperture Radar data and SPOT optical imagery to investigate the co- seismic and post-seismic deformation due to the September 27th 2003, Mw7.3 Altai earthquake, which occurred in the Chuya Basin near the Russia-China-Mongolia border. Based on the SAR and SPOT data, we determined the rupture location and developed a co-seismic slip model for the Altai earthquake. The inferred rupture location is in good agreement with field observations, and the geodetic moment from our slip model is consistent with the seismic moment determined from the tele-seismic data. While the epicentral area of the Altai earthquake is not optimal for radar interferometry (in particular, due to temporal decorrelation), we were able to detect a transient signal over a time period of 3 years following the earthquake. The signal is robust in that it allows us to discriminate among several commonly assumed mechanisms of post-seismic relaxation. We find that the post-earthquake InSAR data do not warrant poro-elastic rebound in the upper crust. The observed deformation also disagrees with linear viscous-elastic relaxation in the upper-mantle or lower-crust. The data can be explained in terms of fault slip in the seismogenic zone, at the margin of areas with high co-seismic slip. Most of the postseismic deformation can be explained in terms of seismic moment release in aftershocks, however, some slip may have occurred aseismically. The observed post-seismic deformation due to the Altai earthquake is qualitatively different from deformation due to other similar-size earthquakes, in particular, the Landers and Hector Mine earthquakes in the Mojave desert, southern California. The observed variations in the deformation pattern may be indicative of different rheologic structure of the continental lithosphere in different tectonically active areas.
G43B-09 INVITED
VLBI, Ring Laser and Gravity Observations for High-Frequency Polar Motion and Universal Time Variations
Very Long Baseline Interferometry (VLBI) is exclusively sensitive to the complete rotation matrix from the terrestrial to the quasi-inertial frame, while ring lasers being sensitive to the instantaneous Earth rotation vector in an Earth- fixed frame, i.e., to polar motion and length of day variations. Additionally, assuming a perfect knowledge of mass attraction variations, superconducting gravimeters (SG) permit to detect Earth rotation variations from the magnitude of the local gravity. The formula of the Sagnac frequency and gravity variation caused by the motion of the Earth's body w.r.t. the instantaneous Earth rotation vector will be presented and compared to VLBI parameterization. The potential of a combination of the three independent types of observables will be examined for sub-diurnal polar motion and universal time variations.
G43B-10
GPS Detection and Modeling of Ionospheric Waves following the 2003 Explosion of the Soufriere Hills Volcano, Montserrat
Volcanic explosions and shallow earthquakes are known to trigger atmospheric waves that propagate at infrasonic speeds in the atmosphere. Because of the exponential decrease of atmospheric density with altitude, the wave's amplitude increases significantly as it propagates upward. Upon reaching ionospheric altitudes, coupling between neutral particles and electrons induces variations of the ionospheric electron density that are detectable by GPS measurements. We used near- and far-field GPS data from Montserrat and the other nearby islands in the Lesser Antilles to examine ionospheric perturbations following the massive dome collapse and explosion of the Soufriere Hills Volcano on July 13th 2003. The ionospheric wave was detected in the GPS- derived integrated electron content (IEC) north of the island, in an area of maximum alignment between theoretical neutral particle motion (from ray tracing) and the Earth's magnetic field, and travels at an apparent velocity consistent with sound speed. Frequency content of the IEC showed peaks at 1 mHz and 4 mHz indicating both a gravity and an acoustic component. The data are consistent with previous observations of atmospheric perturbations after volcanic explosions. To model the acoustic part of the perturbation, we utilized the raytracing equations to follow the path of the wave through the atmosphere. We then coupled the neutral disturbance to the ionosphere by integrating the continuity equation for the charge density with a Chapman electron density distribution. Summation over all ray paths yields a synthetic IEC comparable to the GPS-derived observations.
G43B-11
Volcan de Colima, Mexico: Deformation Monitoring system during the present activity phase 1997-2007
The Volcan de Fuego de Colima (19°30'44"N y 103°37'02"W) is considered to be one of the most active and dangerous volcanoes in Mexico due to its frequent eruptions. It is important to rebound the eruptions of 1585, 1606, 1622, 1690, 1818, 1869, 1890, 1903 and 1913. Around the volcano edifice, a population of more than One Million persons are now living inside a radius of 50 km. The present activity 1997-2007 at Volcan de Fuego de Colima began in November 1997 with a series of seismic swarms and deformation of the summit lava dome. During this period of activity three lava extrusions are occurred: November 1998, May 2001 and September 2004. The deformations surveys at The Volcan de Colima began in 1997 with electronic distance measurements (EDM) carried out with three base stations and nine fixed reflectors located on the summit of the volcano edifice, detecting summit deformation of the extrusion phase of November 1998. In year 1999 began the installation of the electronic tiltmeter network, that send the deformation surveys in real time to the Volcano Observatory (www.ucol.mx/volcan). GPS surveys with a LEICA Double frequency system is carried out since 2006 to determine the variations of the volcano edifice. Here will be presented one analyses of the deformation parameter using different geodesic surveys at the Volcan de Colima.
G43B-12
GPS geodetic constraints on the November 21, 2004 Mw 6.3 earthquake off the northwest coast of Dominica: implications for in situ volatile solubilities and eruption dynamics
Campaign GPS geodesy was initiated in Dominica in 2001. Observations collected by several REU cohorts in
2003, 2004, and 2006 were used to constrain the surface deformation field at 18 sites in 2006. The Mw 6.3
normal faulting event of November 21, 2004, which occurred between the islands of Dominica and Guadeloupe,
affected our GPS interseismic time series. In order to remove the coseismic displacement, we used the Harvard
CMT focal mechanism and aftershock hypocentral locations from the USGS NEIC to define a model fault plane.
Our analysis yields a 35 km fault centered at 15.617 N and 61.709 W with a N33W strike and 55E dip. The
rupture area is 60 by 35 km and the plane projected to the surface is 35 by 34.4 km. From the magnitude of the
main event, the seismic moment was calculated (3.5075x1025 dyne-cm). We assumed constant slip over
the entire fault plane to derive a total finite slip of 9.74 cm.
Our initial fault parameters were used to drive an elastic half space displacement model (DISL) to calculate the
surface displacement from the main shock over a model domain of ~150 km x 150 km. The final slip in the
model was adjusted from 9.7 to 14 cm in order to match an observed coseismic offset in the time series from the
HOUE cGPS site on Guadeloupe. The displacement field from our elastic model indicates northwest motion over
the island of Dominica (south of the fault) and southeasterly motion for Guadeloupe (north of the fault). As
expected, the magnitudes of these displacement vectors increase as distance to the fault decreases.
In order to assess the effect of such a seismic event on nearby magmatic systems in northern Dominica, a
dilatation model was created using Coulomb 2.5. At a depth of 6 km, the model predicts volumetric strain
increase (compression) below the islands of Dominica (Morne Au Diables and Diablotins volcanoes) as well as
Basse-Terre Guadeloupe (Soufriere volcano), with a decrease beneath the fault plane. We conclude that the
dynamic dilatation associated with Mw 6.3 seismic event would increase in situ magmatic volatile solubilities,
and thus would lower the probability of an immediate eruption from a shallow source.
http:comp.uark.edu/~mattioli/Dominica_REU_intro.html
G43B-13
InSAR derived horizontal subsidence gradient as a tool for hazard assessment in urban areas.
Mexico City's subsidence has been recognized for over a century, after the first well battery was drilled to supply water to the rapidly growing city at the end of the XIX century. Consequences of the subsidence process are costly but the economic consequences of subsidence in urban areas is hard to assess due to the fact that their costs are generally factored into yearly maintenance budgets rather than accounting for them as a unique natural disaster. It has thus become increasingly important to assess the extent and magnitude of damage due to ground subsidence in the Mexico City metropolitan area. Periodic InSAR and GPS measurements including data from an 8 site permanent GPS network in Mexico City has enabled us to derive an integrated composite displacement map for Mexico City. This image was used to calculate the horizontal subsidence gradient for areas smaller than 100x100m throughout the city by computing the observed displacement difference divided by the respective pixel to pixel horizontal distance and comparing similarly calculated values from all adjacent pixels in order to select the maximum gradient. Both empiric and analytical approaches can been used to correlate the potential for surface faulting in high horizontal subsidence gradient zones and show that the principal factor for constraining these areas can be successfully determined using a combined Differential InSAR and GPS approach. We present surface faulting risk maps for several cities in central Mexico derived from the horizontal subsidence gradient approach as an example of this technique.
G43B-14
Geodetic And Geological Analysis Of The Tandilia Crust
Keywords: Tandilia-Geoid-Anomaly-Collision-Transamazonic The oldest Precambrian rocks of the south-western Gondwana in South America are cropping out in the Río de la Plata craton, it encompasses the western region of Uruguay, the Martín Garcia island and the Tandilia Ranges in the Buenos Aires Province, Argentina. The Tandil Ranges are the oldest region in Argentina (1.8- 2.2Ga); however, some features of the crust still remain unknown. These rocks evolved during two main events:Transamazonian and Brasilian tectonic cycles. The local and regional gravitational effects were analyzed on gravity and height anomalies. The studied are extended on 400 km x 400 km area which includes three geological units: the Tandilia ranges, and the Claromecó and Salado basins. Due the dependence of gravity and height anomalies with the distance, the seconds are more suitable to analyze the crust interior. For this reason a very precise cuasi geoid model was calculated using the point masses method, from gravity and GPS/leveling data. Taking into account the topography (less than 500 m high hills) and the Bouguer anomalies values (| AB | < 35 mGal), the differences between the cuasi geoid and the geoid will are less than 1 cm. In consequence the geoid undulations (N) were used. The EIGEN-CG01C geopotential model allowed to remove the wavelengths of more than 1000 km of the local geoid (Nobs=N-Neigen). This made it possible to analyze the signals attributed to the structure of Tandilia, and the Claromecó and Salado basins influence on it. In Tandilia geophysical, geodetic and geologic analysis allowed to postulate an isostatic compensated Airy model (Nisost) with a 42 km thick crust (2.84 gr/cm3) resting on a 3.24 gr/cm3 density mantel. Residuals geoid distribution (Nobs-Nisost) shows a tendency to the eastern edge of the ranges. This anomaly is interpreted as the presence of an upper crust excess of mass, which, from Olavarría and Azul hills (north of Tandilia) it extends to the southwest running mainly along the eastern edge of the ranges until the continental platform. This anomaly is attributed to a basic-ultrabasic tectonic slab, a relict of a suture (oceanic bottom and astenosphere) from a continent-continent collision (Transamazonian orogeny). This collisional model was previously based on the presence of wide areas of gneisses, migmatites and granitoids (leucogranites), of sub volcanic and lava flows, of a polifase metamorphic-deformational style, of swarms of pre-metamorphic belts, and of a strong piling up of the crust, associated to thrusting and transcurrence; as well as minor lenses of ultrabasic rocks, that seem to also be part of the suture as a result of a "mantel pinching" during collision.