Spatio-temporal evolution of aseismic slip along the Haiyuan fault, China: Implications for fault frictional properties

https://doi.org/10.1016/j.epsl.2013.07.020Get rights and content

Abstract

We use 20 years of Synthetic Aperture Radar acquisitions by the ERS and Envisat satellites to investigate the spatial and temporal variations of strain rates along the 35-km long creeping section of the Haiyuan fault, at the north eastern boundary of the Tibetan plateau. We then use the derived displacements to infer the faultʼs frictional properties and discuss the relationship between creep and the seismic behavior of the fault. Located in between a millennial seismic gap and the 1920 M8 surface rupture trace, this section has an average creep rate of 5±1mm/yr, about the interseismic loading rate. The comparison of average surface velocity profiles derived from SAR interferometry across the creeping section reveals a creep rate increase and/or a creep migration to shallower depth between the 1990s and the 2000s. We apply a smoothed time series analysis scheme on Envisat InSAR data to investigate the creep rate variations during the 2004–2009 time period. Our analysis reveals that the creep rate accelerated in 2007, although data resolution does not allow to better constrain the onset of creep acceleration and its amplitude. Both decadal and short term transient behaviors are coeval with the largest earthquakes (M45) along the fault segment in recent years. From the precise mapping of the surface fault trace, we use the fault strike variations and the Mohr circle construction to compute the along-strike distribution of the friction coefficient along the creeping segment and compare it with the observed distribution of the creep rate. We find that the creep rate scales logarithmically with the friction coefficient, in agreement with the rate-and-state friction law in a rate strengthening regime. The estimated value of δμ/δlogV2×103 indicates that the earthquakes occurring along the creeping section cannot be the cause for a significant change in the overall segmentʼs creep rate and that the recorded micro seismicity is most likely creep-driven. Finally, given the size and frictional properties of the creeping section, we estimate, based on previous models of dynamic rupture simulations, a 0–20% probability for a rupture to break through this section. Together with the geometrical configuration of the Haiyuan fault, these results suggest that the creeping segment may act as a persistent barrier to earthquake propagation.

Introduction

Constraining the location and size of locked asperities along major faults is a key to a better understanding of their seismic behavior. Modern geodetic techniques, such as GPS or SAR interferometry, allow us to map the degree of locking along subduction interfaces (e.g. Mazzotti et al., 2000, Chlieh et al., 2011, Métois et al., 2012) or major strike-slip faults (e.g. Ryder and Bürgmann, 2008, Jolivet et al., 2012, Lienkaemper et al., 2012), leading to the identification of large locked asperities and estimations of the build-up rate of slip deficit to be released in future earthquakes. Locked asperities are surrounded by areas where slip mostly results from aseismic creep. Observations and modeling studies show that earthquakes tend to nucleate near the transition between locked and creeping patches and that creeping patches act as barriers to the propagation of seismic ruptures (e.g. Lapusta et al., 2000, Cattin and Avouac, 2000, Barbot et al., 2012). The numerical modeling of a frictional interface reveals that the barrier effect of an aseismic patch depends on its size and frictional properties (Kaneko et al., 2010). It is therefore of great importance to identify creeping sections along faults and determine their lateral extension as well as their mechanical properties.

Creeping patches have been identified along several major strike-slip fault systems, such as the San Andreas Fault system (e.g. Rogers and Nason, 1971, Lisowski and Prescott, 1981, Ryder and Bürgmann, 2008), the North Anatolian fault system (Ambrasey, 1970, Çakir et al., 2005) and the Leyte fault, in the Philippines (Duquesnoy et al., 1994). Creep has been detected as well along normal faults (Doubre and Peltzer, 2007), thrust fault systems (e.g. Lee et al., 2001, Champenois et al., 2012) and subduction megathrusts (e.g. Suwa et al., 2006, Chlieh et al., 2008, Moreno et al., 2010, Loveless and Meade, 2011b, Ozawa et al., 2011). The time evolution of creep can be monitored using repeated field observations (e.g. Steinbrugge et al., 1960), creepmeters and more recently, GPS (e.g. Rolandone et al., 2008) and InSAR (e.g. Bürgmann et al., 2000, Funning et al., 2007, De Michele et al., 2011). Various temporal behaviors have been observed including steady creep (e.g. Titus et al., 2006), potentially initiated by post seismic afterslip (Çakir et al., 2012), quasi-periodic seasonal creep rate variations (e.g. Chang et al., 2009) or episodic transient creep (e.g. Lienkaemper et al., 1997, Lienkaemper et al., 2012). Such variability in the aseismic slip behavior affects the slip budget along faults, hence their seismic potential. Therefore, one needs not only to identify creeping patches but also to characterize their temporal evolution.

In this study, we focus on the Haiyuan Fault System (HFS), a 1000 km-long left-lateral system bounding the Tibetan plateau to the North-East (Gaudemer et al., 1995). Using a stack of sparse ERS InSAR data, Cavalié et al. (2008) revealed the existence of a narrow zone of high velocity gradient across the central section of the Haiyuan fault. They proposed that shallow creep was occurring in the brittle upper crust. From the time series analysis of densely sampled Envisat InSAR data, Jolivet et al. (2012) precisely identified a 35 km-long creeping section along the Haiyuan fault, also characterized by an intense seismic activity (Fig. 1). This section is located west of a pull-apart basin that marks the western end of the 240 km-long rupture of the 1920, M8, Haiyuan earthquake (Zhang et al., 1987, Gaudemer et al., 1995). It also corresponds to the eastern termination of a 250 km-long millennial seismic gap, referred to as the Tianzhu seismic gap, likely at the end of its interseismic period and prone to generate M8 earthquakes (Gaudemer et al., 1995, Liu-Zeng et al., 2007). Jolivet et al. (2012) thus showed that this creeping segment is located in between two locked fault sections at different stages of their earthquake cycle, next to a major geometric discontinuity along the fault that is thought to have played a significant role in past earthquakes rupture propagation (Liu-Zeng et al., 2007).

The shallow creep rate is estimated to be 5±1mm/yr, on average, over the 2003–2009 time span of the Envisat data (Jolivet et al., 2012). This value is consistent with the present-day loading rate, determined using GPS or InSAR data (Gan et al., 2007, Cavalié et al., 2008, Loveless and Meade, 2011a). It is also in agreement with the lower bounds of the estimated Holocene slip rates near the eastern end of the Tianzhu gap ranging between 3 and 12 mm/yr (Zhang et al., 1988, Lasserre et al., 1999, Li et al., 2009). The shallow creep rate spatial distribution, averaged over the 2003–2009 period, locally peaks up to 8 mm/yr, which overcomes the present-day tectonic loading rate at depth. This is interpreted as the signal of a temporally variable creep rate along that segment, at least during the period of observation (Jolivet et al., 2012).

In this study, we further investigate the temporal evolution of creep at the eastern end of the Tianzhu seismic gap and its relationship with the recorded seismic activity, at different time scales. We first analyze the creep rate variations at the decadal scale, comparing average surface velocity from ERS (1990s) and Envisat (2000s) data. We then evaluate the short-term evolution of the spatio-temporal distribution of surface creep using Envisat data, spanning the 2004–2009 period. Finally, we use these data to investigate the frictional properties along the fault and discuss the implications of our results for the earthquake cycle and the seismic activity of the creeping section.

Section snippets

Decadal creep rate variations and seismic activity

Comparing results from Cavalié et al. (2008) and Jolivet et al. (2012) suggests temporal variations between the two observation periods (Fig. 2). Cavalié et al. (2008) produced a profile across the Haiyuan fault of the horizontal, fault-parallel velocity, averaged over the 1993–1998 period, by stacking ERS interferograms in the overlapping area of tracks 333 and 061 (Fig. 2b). Jolivet et al. (2012) processed Envisat ASAR data using a time series analysis method on the same track 333 and 061 to

Data processing and time series analysis

We analyze Envisat ASAR data from ascending tracks 240 and 469 and descending track 061, spanning the 2004–2009 period. For these three tracks respectively, 25, 21 and 36 acquisitions are combined into 74, 48 and 131 interferograms, using the ROI_PAC software (Rosen et al., 2004) associated with the NSBAS chain, that enhances coherence over areas with steep topography (Doin et al., 2011). The pair-wise selection of interferograms is such that each acquisition is connected to the others by at

Frictional properties of the creeping segment of the Haiyuan fault

Whether a fault creeps aseismically or produces episodic seismic slip events depends primarily on its frictional properties and more particularly on how friction varies with slip rate (e.g. Brace and Byerlee, 1978, Dieterich, 1979, Scholz, 1998). Assuming a uniform stress field, unperturbed by small earthquakes, the along-strike creep rate variations observed along the Haiyuan fault creeping segment could actually result from spatial variations of frictional stresses related to the fault

Conclusion

The Haiyuan fault segment, immediately west of the section that ruptured during the 1920 M8 earthquake, has been creeping continuously over the InSAR observation periods 1993–1998 and 2003–2009. This creeping segment is about 35 km-long and the average slip rate matches approximately the long term tectonic loading rate on the fault. However, both decadal and short-term temporal variations of the surface creep rate can be observed. Spatial variations of the creep rate can be used to assess the

Acknowledgments

The SAR data set was provided by the European Space Agency (ESA) in the framework of the Dragon 2 program (ID 2509 and 5305). This program also supported R. Jolivetʼs work, through the Young Scientist fellowship. Funding was provided by the French “Extraction et Fusion dʼInformation et de Données dʼInterférométrie Radar” program (EFIDIR, ANR, France) and Programme National de Télédétection Spatiale (CNES). Part of G. Peltzerʼs contribution was done at the Jet Propulsion Laboratory, California

References (89)

  • S. Barbot et al.

    Under the hood of the earthquake machine: Toward predictive modeling of the seismic cycle

    Science

    (2012)
  • P. Berardino et al.

    A new algorithm for surface deformation monitoring based on small baseline differential SAR interferograms

    IEEE Trans. Geosci. Remote Sens.

    (2002)
  • W.F. Brace et al.

    Stick-slip as a mechanism for earthquakes

    Science

    (1978)
  • E.E. Brodsky et al.

    New constraints on mechanisms of remotely triggered seismicity at long valley caldera

    J. Geophys. Res.

    (2005)
  • R. Bürgmann et al.

    Earthquake potential along the northern Hayward fault, California

    Science

    (2000)
  • J.D. Byerlee

    Friction of rocks

    Pure Appl. Geophys.

    (1978)
  • Z. Çakir et al.

    Onset of aseismic creep on major strike-slip faults

    Geology

    (2012)
  • B.M. Carpenter et al.

    Frictional behavior of materials in the 3d SAFOD volume

    Geophys. Res. Lett.

    (2009)
  • B.M. Carpenter et al.

    Weakness of the San Andreas fault revealed by samples from the active fault zone

    Nat. Geosci.

    (2011)
  • R. Cattin et al.

    Modeling mountain building and the seismic cycle in the Himalaya of Nepal

    J. Geophys. Res.

    (2000)
  • S.-H. Chang et al.

    Modelling temporal variation of surface creep on the Chihshang fault in eastern Taiwan with velocity-strengthening friction

    Geophys. J. Int.

    (2009)
  • M. Chlieh et al.

    Heterogeneous coupling of the Sumatran megathrust constrained by geodetic and paleogeodetic measurements

    J. Geophys. Res.

    (2008)
  • M. Chlieh et al.

    Interseismic couping and seismic potential along the central Andes subduction zone

    J. Geophys. Res.

    (2011)
  • J.H. Dieterich

    Modeling of rock friction: 1. experimental results and constitutive equations

    J. Geophys. Res.

    (1979)
  • M.P. Doin et al.

    Presentation of the small baseline NSBAS processing chain on a case example: the Etna deformation monitoring from 2003 to 2010 using Envisat data

  • C. Doubre et al.

    Fluid-controlled faulting process in the Asal Rift, Djibouti, from 8 yr of radar interferometry observations

    Geology

    (2007)
  • T. Duquesnoy et al.

    Detection of creep along the Philippine fault: First results of geodetic measurements on Leyte island, central Philippine

    Geophys. Res. Lett.

    (1994)
  • A. Ferretti et al.

    Permanent scatterers in SAR interferometry

    Geosci. Remote Sens.

    (2001)
  • J. Fukuda et al.

    Fault friction parameters inferred from the early stages of afterslip following the 2003 Tokachi-Oki earthquake

    J. Geophys. Res.

    (2009)
  • G.J. Funning et al.

    Creep on the Rodgers creek fault, northern San Francisco bay area from 10 year ps-InSAR dataset

    Geophys. Res. Lett.

    (2007)
  • W. Gan et al.

    Present-day crustal motion within the Tibetan Plateau inferred from GPS measurements

    J. Geophys. Res.

    (2007)
  • Y. Gaudemer et al.

    Partitioning of crustal slip between linked, active faults in the eastern Qilian Shan, and evidence for a major seismic gap, the ‘Tianzhu Gap’, on the western Haiyuan Fault, Gansu (China)

    Geophys. J. Int.

    (1995)
  • R.M. Goldstein et al.

    Radar interferogram filtering for geophysical applications

    Geophys. Res. Lett.

    (1998)
  • S. Guillaso et al.

    Range resolution improvement of airborne SAR images

    IEEE Geosci. Remote Sens. Lett.

    (2006)
  • E.H. Hearn et al.

    Dynamics of Izmit earthquake postseismic deformation and loading of the Düzce earthquake hypocenter

    Bull. Seismol. Soc. Am.

    (2002)
  • W.E. Holt et al.

    Velocity field in Asia inferred from quaternary fault slip rates and global positioning system observations

    J. Geophys. Res.

    (2000)
  • A. Hooper et al.

    Persistent scatterer interferometric synthetic aperture radar for crustal deformation analysis, with application to Volcán Alcedo, Galápagos

    J. Geophys. Res.

    (2007)
  • Y.-J. Hsu et al.

    Spatio-temporal slip, and stress level on the faults within the western foothills of Taiwan: Implications for fault frictional properties

    Pure Appl. Geophys.

    (2009)
  • Y.-J. Hsu et al.

    Frictional afterslip following the 2005 Nias-Simeulue earthquake, Sumatra

    Science

    (2006)
  • Y.-J. Hsu et al.

    Coseismic and postseismic deformation associated with the 2003 Chengkung, Taiwan, earthquake

    Geophys. J. Int.

    (2009)
  • I. Johanson et al.

    Creep and quakes on the northern transition zone of the San Andreas fault from GPS and InSAR data

    Geophys. Res. Lett.

    (2005)
  • R. Jolivet et al.

    Shallow creep on the Haiyuan fault revealed by InSAR

    J. Geophys. Res.

    (2012)
  • Y. Kaneko et al.

    Towards inferring earthquake patterns from geodetic observations of interseismic coupling

    Nat. Geosci.

    (2010)
  • Y. Kaneko et al.

    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.

    (2012)
  • Cited by (0)

    1

    Now at Tectonics Observatory, California Institute of Technology, Pasadena, CA 90125, United States.

    View full text