Summary.html

McCloskey, J., Steacy, S., Cocco, M., Scotti, O., Zabradnik, J., King, G., Nalbant, S., Nostro, C., Chiarluce, L., Baumont, D., Burjanek, J., and F.
Gallovic

Real time estimation of spatial aftershock distributions

j.mccloskey@ulster.ac.uk

McKernon, C.; Ian Main

Earthquake triggering in regional earthquake catalogues as a function of distance and time.

Conor.McKernon@glg.ed.ac.uk

Mignan, A.; David Bowman and Geoffrey King

Accelerating Moment Release (AMR) before large earthquakes: The Stress Accumulation model (SAM) versus the Epidemic-Type Aftershock Sequence model (ETAS)

king@ipgp.jussieu.fr

Nostro, C., L. Chiaraluce, M. Cocco, D. Baumont

Coulomb stress changes caused by repeated normal faulting earthquakes during the 1997-1998 Umbria-Marche (central Italy) seismic sequence

nostro@ingv.it

Ogata,Yosihiko

Detection of Anomalous Seismicity as a Stress Change Sensor

ogata@ism.ac.jp

Olig, S.; Ivan Wong, and Patricia Thomas

Time-dependent probabilistic seismic hazard analyses along the Wasatch Front, Utah: The need for longer paleoseismic records

Susan_Olig@URSCorp.com

Oncel, A.; O.Aydan

Implications of GPS-derived displacement and stress rates on the 2003 Miyagi-hokubu earthquakes

ali-oncel@aist.go.jp,

Ozalp, S.; Omer Emre and Ahmet Dogan

The last two faulting events on Gemlik segment of the southern strand of North Anatolian Fault Zone, NW Turkey

ozalps@mta.gov.tr

Parotidis M., Shapiro S.A., Rothert E.

Pore pressure relaxation due to ascending magmatic fluids: a triggering mechanism for intraplate earthquake sequences

mparotid@hotmail.com

Parsons, T.

Influence of stress-transfer on earthquake probability calculations

tparsons@usgs.gov

Roth, F.

Three-dimensional Coulomb stress transfer of the ruptures planes of the August 1999 Izmit earthquake on the November 1999 Duzce earthquake

roth@gfz-potsdam.de

Rundle, J.B.; Paul B. Rundle , Andrea Donnellan, Don Turcotte and W Klein

Variation, Recurrence and Correlation in Topologically Realistic System-Level Earthquake Stress-Evolution Simulations

jbrundle@ucdavis.edu

Savage, H.; Chris Marone

The Effects of Transient Stressing on Strength and Stability of Laboratory Faults

hsavage@geosc.psu.edu

Shcherbakov, R.; Donald L. Turcotte, and John B. Rundle

Stress transfer and aftershocks

Turcotte@Geology.ucdavis.edu

Steacy, S., Nalbant, SS, and J. McCloskey

On what planes do triggered aftershocks occur?

s.steacy@ulster.ac.uk

Stein, R.; Shinji Toda, Keith Richards-Dinger

Forecasting the evolution of seismicity in southern California: Animations built on earthquake stress transfer

rstein@usgs.gov

Tanaka, S.; Masakazu Ohtake, and Haruo Sato

Spatio-temporal variation of the tidal triggering effect on earthquakes related to the occurrence of large earthquakes

tanaka@zisin.geophys.tohoku.ac.jp

Tiampo, K.F.; J.B. Rundle, W. Klein, and J. Halliday

The pattern informatics (PI) technique as a measure of stress change

ktiampo@seis.es.uwo.ca

Trotta,Julie E. and Terry E. Tullis

Independent Analysis of the Load/Unload Response Ratio Indicates Little Potential for Earthquake Prediction

jtrotta@email.brown.edu

Do triggered earthquakes occur on optimally-oriented planes?
Greg Anderson (UNAVCO) and Jeanne Hardebeck (USGS)
Many stress transfer studies are based on static stress changes computed on optimally-oriented fault planes, which are those whose orientation and sense of slip combine to give the maximum static stress changes from a given mainshock. Fundamentally, such studies rely on two coupled assumptions: (1) earthquakes triggered by static stress changes are most likely to occur on optimal planes; and (2) static stress changes on the fault planes of triggered earthquakes are well-approximated by the stress changes on the optimal planes. Together, we refer to these assumptions as the optimally-oriented plane (OOP) hypothesis. We test the OOP hypothesis using aftershocks of the 1992 Joshua Tree-Landers-Big Bear earthquake sequence in southern California. We compute the orientation and sense of slip for the optimal plane for each event given the event location, a mainshock finite-fault model, Coulomb friction parameter, and background stress field; we also compute uncertainties in the optimal plane given uncertainties in the input parameters. We then compute the percentage of events with compatible optimal planes and focal mechanisms, and the percentage of events with compatible stress change estimates on the optimal and observed focal mechanism planes. Finally, we use a bootstrap resampling technique to estimate the significance of our observations and thus the confidence with which we can accept or reject the OOP hypothesis.
return to list

Upper Mantle Strength Contributions to Crustal Fault Location and Style in California and Western Nevada
Glenn Biasi
Assessing the mantle contribution to stress transfer and total lithospheric strength has been difficult and controversial. Recent modeling of post-seismic relaxation in GPS measurements for Landers tends to confirm a mantle role, but only locally contrains it. The larger problem of mapping results beyond the Mojave remains. Uppermantle tomography provides an alternative means of mapping upper mantle strength. Briefly, mechanisms that increase mantle velocity also tend to strengthen it. Temperature and composition are most important in subsolidus environments. The velocity structure in map view at shallow mantle depth may be interpreted as a map of the relative strength of the mantle contribution. Crust above high velocity mantle lithosphere should deform differently than crust with no support. This model makespredictions that reasonably correlate with fault location and style in California and western Nevada. Spatial variations of strength in the mantle lithosphere would modulate the distance over which stresses interact and might provide a mechanism for longer range interaction of stresses than would be inferred from crustal mechanisms alone.
return to list

Bayesian Inference for the Assessment of Aftershock Hazards
Chen,Yuh-Ing; Ma, Kuo-Fong; Hong, Chien-Hwei
In this paper, we show how to make the Bayesian inference based on the Markov Chain Monte Carlo (MCMC) method for near real-time evaluating the aftershock hazard after, for example, the Chi-Chi earthquake (M=7.3, 1999/9/20UT) in Taiwan. This inference requires current, but limited aftershocks, as well as the hazard information from previous aftershock sequences. We then apply the gridding technique to portray the spatial aftershock hazards based on the Bayesian analysis results. Finally, we investigate the possible relation between the Bayesian version of aftershock hazards and the related stress change due to the Chi-Chi earthquake.
return to list

Mechanical coupling and fault interaction modeled with 3D finite element: an application to Landers - Hector Mine (Southern California) fault systems
S. Cianetti, C. Giunchi and M. Cocco.
We focus on 3D finite element modeling of the Landers-Hector Mine sequence, aiming to understand how the regional stress is able to load pre-existing fault planes. We investigate the mechanical coupling between the two earthquakes evaluating the Coulomb stress perturbations taking into account the complexities associated to the fault plane geometry and rheological heterogeneities in the crust.
return to list

Triggered events in conjugate fault system in the westward-escape tectonic regime of Turkey: The January 27 Pulumur (Mw: 6.1) and May 1 Bingol earthquakes in 2003
Omer Emre, Erdal Herece, Ahmet Dogan, Rukiye Cƒ±plak, Volkan Ozaksoy, Oktay Parlak and Selim Ozalp
Two moderate earthquakes occurred in the Bingol-Karlyova-Erzincan ~ 30 angle (BKEA), at the eastern corner of the Anatolian block, which is characterized by westward escape. The BKEA is located between North (NAF) and East (EAF) Anatolian faults west of the Karlyova triple junction. It is bounded by large step-over structures: the Erzincan basin on the NAF and Gokdere stepover on the EAF. A conjugate active fault pattern includes NE and NW trending faults parallel to the NAF and EAF. The NW and SE trending faults are right-lateral and left-lateral, respectively. The January 27, 2003 Pulumur Earthquake (Mw:6.1) occurred at the northwest corner of the BKEA. Ten years ago, 13 March 1999, the Erzincan earthquake (Ms: 6.8) occurred at the east of the Erzincan basin and triggered, two days later, the Pulumur Earthquake (Mw:5.8) on Dagyolu fault, which is parallel to the NAF. The 27 January 2003 Pulumur earthquake occurred on the Pulumur fault, which is conjugate to the Dagyolu fault. The Pulumur fault is a left-lateral, north-east trending, and 20 km in length. The January 2003 event might be triggered by 1992 Erzincan earthquake. The 01 May 2003 Bingol earthquake occurred at the southeastern corner of the BKEA. Before the last event, the May 22 1971 Bingol earthquake (Ms:6.8) occurred on the Goynuk segment of the EAF and produced surface rupture. The Bingol-Karakocan and Sudugunu faults are conjugate to the 1971 rupture of EAF. The earthquake in the region, the 01 May 2003 Bingol earthquake, was generated by the Sudugunu fault but did not rupture the surface. However, many landslides and lateral spreading were observed along this fault 20 km-long and N500 W trending fault. The 2003 Pulumur and Bingol earthquakes were close in time and space to the 1992 Erzincan and 1971 Bingol earthquakes, respectively. The series of earthquakes during the last 30 years suggest that triggering is common in the conjugate fault pattern of the BKEA. Therefore, it can be speculated that the 2003 events raised the earthquake potential on the adjacent faults.
return to list

Hidden Markov Models for Pattern Recognition and Classification of GPS, Seismicity, and Broadband Seismic Data
Robert Granat, Robert Clayton, Sharon Kedar
We describe a method that employs hidden Markov models to analyze geophysical time series data. We apply our method to several sources of data collected in Southern California: GPS measurements of crustal deformation, seismicity records, and broadband seismographs. The method works in the absence of a proiri information about the data, is capable of classifying modes in the time series and deriving statistical relationships between events and behaviors.
return to list

Detection of Anomalous Seismicity as a Stress Change Sensor
Yosihiko Ogata (Institute of Statistical Mathematics, Tokyo)
I will discuss seismicity changes (quiescence and activation), relative to the predicted seismicity rate by the fitted ETAS model, in various regions in and around northern Japan, which are consistent with the co-seismic stress changes due to the recent strong earthquakes. These lead us to a summarized observation that even small sizes of CFF changes down to the order of millibars can trigger such activation or lowering of micro-seismicity, which is also supported by the Dieterich's seismicity rate-change equation [Dieterich, 1994]. Thus, we expect that aftershock activity and general seismic activity relative to the modeled rates is sensitive enough to measure a slight stress change. Likewise, an assumed precursory aseismic slip during the summer of 1996 within or around the rupture area of the 2003 Tokachi earthquake of M8.0 can consistently explain the simultaneously occurred relative activation of seismicity in inland of northern Tohoku District, relative activation of the secondary aftershock activity of the 1994 Sanriku-Haruka-Oki earthquake of M7.8, and relative quiescence of aftershock activity of the 1993 Hakkaido Nansei-Oki earthquake of M7.7.
return to list

Detecting fluid signals in seismicity data through statistical earthquake modeling
Sebastian Hainzl and Yosihiko Ogata
We show that the Epidemic Type Aftershock (ETAS) model is an appropriate tool to extract a primary fluid signal from complex seismicity pattern. We analyze a large earthquake swarm occurred in the year 2000 in Vogtland/NW-Bohemia, Central Europe. By fitting the stochastic ETAS model, we find that stress triggering is dominant in creating the seismicity patterns. This explains the observed fractal interevent-time distribution. External forcing identified with pore pressure changes due to fluid intrusion is found to directly trigger only about 1% of the total activity. However, a temporal deconvolution unveils a pronounced fluid signal initiating the swarm. By analyzing model simulations in which earthquakes are triggered by fluid intrusion as well as stress changes on a fault plane embedded in a 3d-elastic half-space, we find that the proposed deconvolution procedure is able to reveal the underlying pore pressure variations.
return to list

Dynamic Stress Triggering within Earthquakes: Predicting magnitudes of large earthquakes on the Hayward fault, California and the Alpine fault, New Zealand
R.A. Harris, D. Eberhart-Phillips, A. McGarr, D. Oglesby/B. Aagaard, S. Day
This paper will use spontaneous (fully dynamic, including inertia) rupture simulations of scenario large earthquakes on the Hayward fault, California and the Alpine fault, New Zealand to determine how strength variations on these two faults will control the sizes of their upcoming large earthquakes. This is a stress triggering study since the earthquakes only continue to propagate if the stress waves generated during the events are sufficient to trigger ongoing rupture propagation.
return to list

The probability of jumping stepovers in vertical strike-slip faults
R.A. Harris, S. Day
This long ago promised work will finally allow earthquake probability folks to include the probability of multi-segment rupture in earthquake forecast scenarios. The specific study will investigate the statistical likelihood of an earthquake on a vertical strike-slip fault cascading across a stepover to continue onto a neighboring fault. The methodology is 3D spontaneous rupture simulation and this is a stress triggering study of multi-fault triggering.
return to list

Effect of small earthquakes for static stress changes and earthquake triggering
Helmstetter, A.; Yan Kagan and David Jackson
We evaluate the relative importance of small and large earthquakes for static stress changes and earthquake triggering, which depends on the spatial distribution of seismicity. The stronger the spatial clustering is, the larger is the influence of small earthquakes for stress changes at the location of a future earthquake. We evaluate the fractal dimension D of hypocenters using a relocated earthquake catalog. The value of D=1.6 is in good agreement with the exponent alpha=0.8 of the scaling of the number of aftershocks with the mainshock magnitude, and with the hypothesis that aftershocks are due to static stress changes. The value of D<2 implies that small earthquakes are collectively more important than larger ones for static stress changes and earthquake triggering.
return to list

Remotely Triggered Earthquakes in Diverse Tectonic Settings: Evidence for Brittle Faults?
Susan E. Hough
Since 1992, remotely triggered earthquakes have been identified from large (generally M>7) earthquakes in California as well as in other regions. These events occur predominantly, although not exclusively, in active geothermal/volcanic regions, leading to theories that the earthquakes are triggered when passing seismic waves cause disruptions in magmatic or other fluid systems. In this paper I focus on observations of remotely triggered earthquakes outside of active geothermal and volcanic regions. Recent results from California suggest that triggered earthquakes are low stress drop events that occur on especially weak faults--or on especially weak segments of otherwise strong faults. This, combined with the absence of triggered earthquakes along most of the San Andreas fault, provides support for a brittle fault model: i.e., that faults such as the San Andreas are statically strong but dynamically weak. I summarize evidence that remotely triggered earthquakes also occur in mid-continent and collisional zones. This evidence is derived from both careful analysis of historic earthquake sequences and from instrumentally recorded M5-6 earthquakes in eastern Canada. Preliminary results suggest that remotely triggered earthquakes occur preferentially at the distances at which SmS waves are known to significantly amplify ground motions. This lends further support to the conclusion that, even at distances as small as ~100 km, dynamic stress changes control the occurrence of triggered events. I discuss two alternative explanations to account for the occurrence of remotely triggered earthquakes in intraplate settings: 1) that, like interplate remotely triggered earthquakes, they occur at local zones of weakness along otherwise strong faults, or 2) that intraplate triggered earthquakes occur in zones of local stress concentration.
return to list

Stress in southern California,1850-2003
Kagan, Y. Y., D. D. Jackson, and Z. Liu
We compute the incremental stress tensor in the upper crust of southern California as a function of time since 1850, and we compare observed seismicity with the estimated stress increment at the time of each earthquake. We model crustal deformation using updated geodetic, especially GPS data, and geologically determined fault slip rates. We consider the stress increments from previous earthquakes, and the aseismic tectonic stress, both separately and in combination. The locations and mechanisms of earthquakes are best correlated with the aseismic shear stress, which grows at a constant rate in our model. In general, correlations between normal stress and earthquakes often change sign if we modify even slightly the initial parameters of our computations (like starting date of the catalog, exclusion of earthquake with closely spaced epicenters, use of point vs distributed sources). Although the correlation of tectonic stress with earthquake triggering is robust, other results are unstable apparently because there are relatively few earthquakes in the catalog.
return to list

Earthquake slip distribution
Kagan, Y. Y.
I consider statistical earthquake slip distributions as can be inferred from seismic moment-frequency relations and geometrical scaling for earthquakes. Using various assumptions on temporal earthquake occurrence, these distributions can be used to evaluate accuracy of geologic fault slip determinations and estimate uncertainties in long term earthquake patterns based on paleoseismic data.
return to list

Accelerating Moment Release (AMR) before large earthquakes: The Stress Accumulation model (SAM) versus the Epidemic-Type Aftershock Sequence model (ETAS)
Arnaud Mignan, David Bowman and Geoffrey King
The Epidemic-Type Aftershock Sequence (ETAS) model has been proposed as an explanation for the widely observed phenomenon of accelerating moment release (AMR) before large earthquakes. This model is based on the idea that large earthquakes are the end result of a period of self-organization in the background stress field (i.e. smaller earthquakes). In contrast to this concept is the Stress Accumulation model (SAM), which treats the evolution of seismicity before a large earthquake as the result of loading of the main fault primarily by creep on an extension of the fault at depth. While both of these models predict accelerating seismicity before the mainshock, they make different predictions of the spatial distribution of this activity. We examine the statistical significance of AMR for the spatial distributions for the two models by comparing them to random catalogues. The behaviour predicted by SAM is supported with a probability greater than 99%. Only for two events Coalinga and Hector Mine can the ETAS model possibly be invoked. Our study does not suggest that the stress coupling mechanism that underlies the ETAS model is incorrect; merely that it is not the main cause of AMR.
return to list

Earthquake Triggering and Fault Interaction in a Thrust Fault System: the 1999 Chi-Chi, Taiwan, Earthquake
Kuo-Fong Ma, Chung-Han Chan, Ross S. Stein
To understand earthquake triggering and possible fault interaction in a thrust fault system, we calculate the Coulomb stress change associated with the 1999 Chi-Chi, Taiwan, earthquake. The Coulomb stress change for the ruptured Chelungpu fault shows broad positive distribution of Coulomb stress change both in space and depth. The correlation of the Coulomb stress change with the aftershocks is significant. The triggering of the aftershocks has the Coulomb stress change of less than 1 bar in off-source regions,
especially to the northeast of the fault. The thrust faulting mechanism of the Chi-Chi earthquake triggered the aftershocks distributed in a southeastern direction with strike-slip focal mechanisms in the southern tip of the thrust fault. Even though the Coulomb stress change of the earthquake is up to about 500 bars, the calculated Coulomb stress change for the optimal fault planes of the M>6 aftershocks are mostly about 2-5 bars. It suggests that the large aftershocks are not necessary occur in the region with large Coulomb stress change. Considering the fold-and-thrust faulting system in Taiwan, the Coulomb stress change for the fault system associated with the Chi-Chi earthquake shows significant stress relaxation for the thrust faults to the east of the ruptured Chelungpu fault. The imparted positive stress on the Changhua and Chiko faults, to the west and south of the Chelungpu fault, are prominent. Comparison with historical events since 1600, the calculated Coulomb stress change shows strong correlation of the occurrences of the earthquake on the Chelungpu fault to the events on Changhua and Chiko faults in the 19th century. This seems to suggest the possible predictive power of the static stress calculation in estimating occurrences of future large earthquakes.
return to list

Testing Coulomb Stress Transfer Models with Seismicity Rates for the Landers and Kobe Earthquakes
Ellen P. Mallman and Mark D. Zoback
We present work that quantitatively compares the change in Coulomb stress and the long-term seismicity rate of the surrounding region in response to 1992 Landers and 1995 Kobe earthquakes. Seismicity rates are quantified several ways including the beta-statistic, the z-value, and the average number of earthquakes in a given volume. The change in Coulomb stress is calculated assuming both a regionally uniform stress field (as is commonly done) and a realistic regional stress field obtained from focal mechanism inversions and other stress measurements. We quantitatively compare the results from the various methods of calculating rate change and the change in Coulomb stress.
return to list

Real time estimation of spatial aftershock distributions.
McCloskey, J., Steacy, S., Cocco, M., Scotti, O., Zabradnik, J., King, G., Nalbant, S., Nostro, C., Chiarluce, L., Baumont, D., Burjanek, J., and F. Gallovic
The observed strong relation between Coulomb stress changes and the spatial distribution of aftershocks suggests that the real-time estimations of likely aftershock distributions may be feasible. Here we report the results of the PRESAP project which investigated whether the science is sufficiently well advanced for such assessment: in particular we discuss the first order effects on Coulomb stress maps, report the results of a pseudo-real-time test, present a recipe for making real time stress maps, and recommend specific scientific and technological improvements to move this hazard estimation from the feasible to the practical.
return to list

Three-dimensional Coulomb stress transfer of the ruptures planes of the August 1999 Izmit earthquake on the November 1999 Duzce earthquake
F. Roth
One criterion for stress triggering of earthquakes is the so-called Coulomb failure function. It incorporates a combination of the shear stress along a potentially active fault and the normal stress on this fault i.e. the "Coulomb stress". The shear stress has to be high enough to overcome friction and the normal stress should favour Unclamping of the fault. During the last years, this approach was mainly used assuming a meta-stable condition of the earth’s crust, therefore: If the change in Coulomb stress by the initial event is positive, failure through another earthquake is expected. We followed this assumption. However, the limitation so far was that the rupture planes were restricted to vertical strike-slip faults. In reality, the rupture planes of two events are often inclined and differ in strike and in dip. Here, we present results for this more general case, applied to 2 couples of earthquakes (i) in the Fiji-Tonga subduction zone and (ii) on the North Anatolian fault zone (NAFZ). The Fiji-events occurred on August 19, 2002, the first at 11:01 and the second at 11:08 UTC. Their magnitude was determined as Mw=7.5 and Mw=7.7, respectively, with a depth offset of less than 100 km and a hypocentral distance of about 290 km. The NAFZ-events occurred August 17 and November 12, 1999, with magnitudes Mw 7.4 and 7.1 on adjacent rupture planes with different strike and dip. We present an analysis of the Coulomb stress changes at the site of the 2nd event due to stress changes by the 1st event. Doing so, we varied the relative position, the size of both rupture planes and the fault plane solutions of both events from different catalogues.
return to list

On what planes do triggered aftershocks occur?
Steacy, S., Nalbant, SS, and J. McCloskey,
Coulomb stress maps are produced by computing the tensorial stress perturbation due to an earthquake rupture and resolving this tensor onto planes of a particular orientation. Following Stein et al. (1992) and King et al. (1994), such planes are generally considered to be "optimally oriented", in other words the modeller uses the regional stress and co-seismic stress change to compute the orientation of planes most likely to fail and resolves the co-seismic stress onto these orientations. This practice implicitly assumes that faults capable of sustaining aftershocks exist at all orientations; an assumption contradicted by the observation that aftershock focal mechanisms have strong preferred orientations consistent with mapped structural trends. Here we systematically investigate the best planes onto which stress should be resolved by quantitatively comparing observed aftershock distributions with stress maps based on optimally oriented planes (2D and 3D), mainshock orientation, and regional structural trend for a number of well constrained earthquakes.
return to list

Time-dependent probabilistic seismic hazard analyses along the Wasatch Front, Utah: The need for longer paleoseismic records
Olig, Susan, Ivan Wong, and Patricia Thomas
This paper will discuss the methodolgy and results of PSHAs conducted along the Wasatch Front that compared both time-dependent and Poisson models for earthquake occurence on the Wasatch fault. We incorporated the most recent data that extends the paleoseismic record back 15 ka on some fault segments to develop time-dependent recurrence intervals for a lognormal renewal model using the approach of the 1999 Working Group on California Earthquake Probabilities. Recurrence intervals for the extended record vary considerably and results show that these variations significantly impact the hazard, increasing ground motions by over 60% or decreasing them by 20%, depending on how variations in recurrence are incorporated and what assumptions are made about about earthquake perodicity. These results clearly show the importance of obtaining more extended paleoseismic records to better understand just how periodic fault behavior is, and what are the causes for the large variations in recurrence that are seen on many faults. If we are to move forward with using time-dependent models in PSHAs, we need answers to these questions.
return to list

Implications of GPS-derived displacement and stress rates on the 2003 Miyagi-hokubu earthquakes
A. Oncel, O.Aydan
Earthquake prediction is an important field of research for years, that resulted some success on long-term and intermediate prediction but short-time prediction is still unpredictable . Thus, some earthquake prediction projects such as the Tokai Earthquake Project in Japan, the Parkfield Earthquake project in USA have been recently undertaken to progress on short-term prediction. One of the challenges in earthquake prediction is associated to have little knowledge about the elasto-visco-plastic characteristics and structure of the crust and their spatial distribution and complexity of loading conditions.
If the stress state and the yielding characteristics of the earth's crust are known at a given time, one may be able to predict earthquakes with the help of some mechanical, numerical and instrumental tools. The stress rates derived from the GPS deformation velocity measurements can be effectively used to locate the areas with high seismic risk. Thus, daily variations of derived strain-stress rates from dense GPS networks in Japan and USA may provide a high quality data to understand the behaviour of the earth's crust preceding earthquakes.
In this article, we use an interpolation technique of finite element method proposed by Aydan [Aydan, 2000; Aydan, 2003] to compute the strain rate and consequently stress rate in the tangential plane to the surface of the earth's crust from crustal deformations, which is briefly presented. This method is then applied to the GPS measurements in the Northern part of Miyagi Prefecture, where recently large earthquakes with a magnitude of 7.0 (May 26, 2003) and 6.2 (July 26, 2003) occurred. The stress rates for the area in the close vicinity of the epicenter of M6.2 earthquake are computed and compared with seismic activity, and their implications are discussed. In addition, the general tendency of deformation pattern obtained from GPS measurements are discussed with that expected from the tectonic model of the prefecture.
return to list

The last two faulting events on Gemlik segment of the southern strand of North Anatolian Fault Zone, NW Turkey
Selim Ozalp, Omer Emre and Ahmet Dogan
The aim of the paper is to discuss of the earthquake potential of the southern strand of North Anatolian Fault Zone in the Marmara Region based on the paleoseismic data. All modelling studies based on GPS data were focused on the northern strand of the NAF after the 1999 Izmit earthquake. However our paleoseismic results along Gemlik segment of the southern strand reveal that high earthquake potential of the southern strand as much as northern strand. In this paper, we would like to discuss the earthquake hazard along the southern strand of the North Anatolian Fault Zone.
return to list

Pore pressure relaxation due to ascending magmatic fluids: a triggering mechanism for intraplate earthquake sequences
Parotidis M., Shapiro S.A., Rothert E.
1. The triggering mechanism of pore pressure diffusion: theory, seismic signatures, numerical modeling. 2. Case study: earthquake swarms in Vogtland/NW-Bohemia; a. geologic, geophysical, and tectonic environment; b. seismic activity; c. data analysis of the year 2000 events; d. determination of seismic signatures; e. numerical modeling of spatio-temporal signatures; f. comparison of our results (hydraulic and strength parameters of seismogenic crust) with other studies. 3. Discussion.
return to list

Influence of stress-transfer on earthquake probability calculations
T. Parsons
This is a paper that would explore the range of possible answers one might get for a given set of input parameters. A broad range of
recurrence intervals can fit paleoseismology or historical cats. COV is a major issue. In stress transfer we have clock-change vs. interval changes, uncertainty in loading rates, slip models, dynamic v. static etc. And of course transients. Ultimately I hope to find out what conditions stress transfer broadens the range of answers, and when is it overlapped by all the other uncertainty.
return to list

Variation, Recurrence and Correlation in Topologically Realistic System-Level Earthquake Stress-Evolution Simulations
John B. Rundle, Paul B. Rundle , Andrea Donnellan, Don Turcotte and W Klein
The problem of earthquake forecasting within the context of model-based methods is usually approached using numerical simulations. Along with data assimilation technologies, simulations can be used to formulate ensemble forecasting techniques. We focus on new, topologically realistic system-level approaches to the modeling of earthquake fault systems. These models are backslip models, in which the fault system topology remains fixed in time and does not evolve. In addition to laboratory-based friction models, the model includes 650 rectangular fault segments that interact by means of linear elasticity. Inputs to these models arise from field data, and typically include realistic fault system topologies, realistic long term slip rates, and realistic frictional parameters. Outputs from the simulations include synthetic earthquake sequences and space-time patterns, together with associated surface deformation and strain patterns that are similar to those seen in nature. It is clear from the dynamics of the model that stress transfer plays a central role in the fault slip evolution process, and stress shadows and stress halos can easily be seen. By analyzing the statistical physics of the simulations, we can show that that the frictional failure physics, which includes a simple representation of a dynamic stress intensity factor, leads to self-organization of the statistical dynamics, and produces empirical statistical distributions that characterize the activity. An example is the Gutenberg-Richter magnitude frequency distribution obtained from simulations, which is seen to be similar to the corresponding distribution in nature (although modified by the restricted range of spatial scales in the simulations). Subsequent development of simulation-based methods should include model enhancement, data assimilation and data mining methods, and analysis techniques based on statistical physics.
return to list

The Effects of Transient Stressing on Strength and Stability of Laboratory Faults
Heather Savage and Chris Marone
We analyze shear stress response of a laboratory fault due to transient, periodic loading rate oscillations. By varying amplitude and frequency of the oscillations, we simulate transient stress conditions of natural faults exposed to seismic waves or tidal stress and decipher if these oscillations influence the occurrence of unstable behavior. Additionally, we forward model these conditions to determine if my laboratory results are predicted by rate and state friction theory.
return to list

Stress transfer and aftershocks
Robert Shcherbakov, Donald L. Turcotte, and John B. Rundle.
By combining Gutenberg-Richter scaling, Bath's law, and Omori's law a universal scaling law is found. This scaling and the role of stress transfer will be modeled using damage mechanics.
return to list

Spatio-temporal variation of the tidal triggering effect on earthquakes related to the occurrence of large earthquakes
Sachiko Tanaka, Masakazu Ohtake, and Haruo Sato
We observed remarkable change in time-space pattern of the tidal triggering effect on earthquakes that is closely related to the occurrence of large earthquakes with Mw >= 7.5 in several subduction zones. We precisely measured the correlation between the earth tide and earthquake occurrence. The result of statistical analysis indicates that the tidal phase selectivity concentrated in and around the future focal regions for several years preceding the occurrence of large earthquakes.
return to list

The pattern informatics (PI) technique as a measure of stress change
K.F. Tiampo, J.B. Rundle, W. Klein, and J. Halliday
The subject of this paper will be the theoretical underpinnings of the PI index and its use as a measure of anomalous stress change. At least one case study will be examined, the interaction between Parkfield and Coalinga.
return to list

A systematic search for stress shadows after mainshocks
M. Gerstenberger, D. Schorlemmer, S. Wiemer, and J. Woessner
We investigate long-term and long-distance rate changes caused by large mainshocks in the US and Japan. We quantitatively compare seismicity rate changes in the years preceding and following these mainshocks, averaging over volumes of tens of kilometer in radii. In doing so, we are able to resolve with high significance not only rate increases but also rate decreases. Next we design a likelihood ratio based tests to evaluate if these rate changes correlate significantly with predictions made by the static stress changes caused by these events when resolved onto optimal oriented planes. We also assess the probability gain, if any, of the forecasts made based on static stress changes.
return to list

Independent Analysis of the Load/Unload Response Ratio Indicates Little Potential for Earthquake Prediction
Julie E. Trotta and Terry E. Tullis
The Load/Unload Response Ratio (LURR) proposed method of earthquake prediction (Yin et al., 1995) involves looking at the ratio of the number or some measure of the size of small earthquakes that occur during loading of the tidal cycle to that measure of earthquakes that occur during unloading of the tidal cycle, and suggests this ratio should increase before an impending large shock. We calculated LURR values for several large earthquakes in southern California, using the methods and parameters described by Yin et al. (2000) and estimates of the other necessary methods and parameters that were not published. We also calculated LURR values for altered data sets in which we randomized the times of each small earthquake before calculating the tidal stresses, theoretically eliminating any real influence of the tides on LURR. Results thus far do not indicate any statistically significant difference from the calculations utilizing the actual earthquake times and calculations utilizing the randomized times or that the LURR method is a useful tool in earthquake prediction.
return to list