Fault behavior and lower crustal rheology inferred from the first seven years of postseismic GPS data after the 2008 Wenchuan earthquake

Abstract

Long-term and wide-area geodetic observations may allow identifying distinct postseismic deformation processes following large earthquakes, and thus can reveal fault behaviour and permit quantifying complexities in lithospheric rheology. In this paper, the first 7 years of GPS (Global Positioning System) displacement data following the 2008 Mw7.9 Wenchuan earthquake are used to study the relevant mechanisms of postseismic deformation. Two simple models that consider either afterslip or viscoelastic relaxation as the unitary mechanism of the postseismic deformation are tested at first. After analysing the limitations and complementarity of these two separated models, a combined model incorporating the two main mechanisms is presented. In contrast to previous studies, which mostly assume that afterslip and viscoelastic relaxation are independent, our combined model includes the secondary viscoelastic relaxation effect induced by transient Manuscript to be submitted to EPSL 2 afterslip. Modelling results suggest that the middle-to far-field postseismic deformation is mainly induced by viscoelastic relaxation of the coseismic stress change in the lower crust, whereas the near-field displacement is dominantly caused by stress-driven aseismic afterslip. The seismic moment released by the transient afterslip corresponds to an Mw7.4 earthquake, or 25% of that released by the Mw7.9 main shock. With a characteristic decay time of 1.2 years, most afterslip (~ 80%) is released in the first 2 years. Negligible aseismic afterslip is observed in the seismic gap between the Wenchuan and Lushan earthquakes, indicating the locked state of the fault within this segment. The effective lower crustal viscosity of the eastern Tibetan Plateau is estimated to be 2.0 × 10 18 Pa•s, at least two orders of magnitude smaller than that of the adjacent Sichuan Basin. This finding is consistent with previous observations of low seismic velocity and high electrical conductivity in this region, all of which support the assumption that the crustal thickening in the eastern Tibetan Plateau is dominantly caused by ductile lower crustal flow, with important implications for understanding both long-and short-term crustal deformation processes.

A Gravity Study of the Longmenshan Fault Zone: New Insights Into the Nature and Evolution of the Fault Zone and Extrusion‐Style Growth of the Tibetan Plateau Since 40 Ma

“…The aftershocks occurred almost exclusively along the central and northern LFZ with increasing right‐lateral strike‐slip component toward the north (Figure ; Shen et al, ), probably because the main shock released a locking point that prevented the motion of the Songpan Block. Cumulative postseismic surface displacements during the first 7 years after the Wenchuan earthquake on the apposing sides of the Fubianhe Fault show sharp contrast in both their directions and the magnitudes between the Danba and Songpan blocks (Figure , purple vectors; Diao et al, ), consistent with our interpretation.…”

“…Thin blue arrows show generalized GPS measurements of present‐day crustal motions (Zhang et al, ). Thin violet arrows represent observed cumulative postseismic surface displacements during the first 7 years after the Wenchuan earthquake (Diao et al, ). Ages indicate the estimated timing of dominant plateau uplift for a given region, bounded by red dashed lines (Richardson et al, ).…”

“…We therefore envisage that contrasting behaviors between the southern and central‐northern LFZ were driven by the differing motion directions of the Danba and Songpan blocks, separated by the Fubianhe Fault (FBHF), along the eastern margin of the Tibetan Plateau (Figure ; Guo et al, ; Diao et al, ), with a twisted fault plane (Shen et al, ) adjacent to the bounding fault between the two blocks (Figure c). The M7.9 Wenchuan earthquake occurred near the junction between the Fubianhe Fault and the LFZ (Figures a and c), possibly reflecting intense accumulation of seismic energy over such a unique structural juncture and complex fault geometry.…”

Ambient Noise Monitoring of Seismic Velocity Around the Longmenshan Fault Zone From 10 Years of Continuous Observation

“…Froment et al (2013) suggested that the viscoelastic relaxation of the deep crust and/or postseismic slip may be the reason for the delayed seismic velocity change in the period band of 12-20 s. Obermann et al (2014) inferred that velocity change in this period band may be caused by the detachment of a deep fault segment beneath eastern Tibet in relation to the Wenchuan earthquake. Since the depth range (~5-20 km) of the delayed seismic velocity drop in this study is within the rupture zone or aftershock zone of the Wenchuan earthquake (Zhang et al, 2010) and is shallower than the viscoelastic layers beneath the Longmenshan fault zone (Diao et al, 2018), we postulate that this change is the result of postseismic slip along the deep rupture zone of Wenchuan earthquake.…”

Intermediate‐Magnitude Postseismic Slip Follows Intermediate‐Magnitude (M4 to 5) Earthquakes in California

“…The postseismic observations of larger earthquakes were made by a range of researchers over a range of timescales, from days to years after the earthquakes, as listed in Table S3. The large‐event moment ratios come from 35 earthquakes numbered in Figure , from the 1: 2005 Chaman (Furuya & Satyabala, ), 2: 2008 Mogul, NV swarm (Bell et al, ), 3: 1998 San Juan Bautista (Taira et al, ), 4: 2007 Alum Rock, CA (Murray‐Moraleda & Simpson, ), 5: 2007 Ghazaband (Fattahi et al, ), 6: 2004 Parkfield (Freed, ; Langbein et al, ), 7: 2014 South Napa (Amoruso & Crescentini, ; Cheloni et al, ; Floyd et al, ), 8: 2009 L'Aquila (D'Agostino et al, ), 9: 2008 Nima‐Gaize, Tibet (Ryder et al, ), 10: 2000 Iceland (Jónsson, ), 11: 2003 San Simeon (Johanson & Bürgmann, ), 12: 2003 Zemmouri (Cetin et al, ; Mahsas et al, ), 13: 1989 Loma Prieta (Segall et al, ), 14: 1991 Racha, Georgia (Podgorski et al, ), 15: 1999 Hector Mine (Jacobs et al, ), 16: 2003 Altai (Barbot et al, ), 17: 2010 El Mayor‐Cucapah (Gonzalez‐Ortega et al, ), 18: 2011 Van (Dogan et al, ), 19: 1992 Landers (Savage & Svarc, ), 20: 1997 Manyi, Tibet (Ryder et al, ), 21: 2012 Nicoya (Hobbs et al, ; Malservisi et al, ), 22: 1994 Sanriku (Heki et al, ; Melbourne et al, ), 23: 2015 Gorkha (Sreejith et al, ), 24: 2001 Kokoxili, Tibet (Wen et al, ), 25: 1997 Kronotsky (Bürgmann et al, ), 26: 2016 Pedernales (Rolandone et al, ), 27: 2008 Wenchuan (Diao et al, ), 28: 1995 Jalisco (Melbourne et al, ), 29: 2003 Tokachi‐Oki (Miura et al, ), 30: 1995 Antofagasta (Melbourne et al, ; Pritchard & Simons, ), 31: 2015 Illapel (Shrivastava et al, ), 32: 2001 Peru (Melbourne et al, ), 33: 2005 Nias (Hsu et al, ), 34: 2010 Maule (Lin et al, ), and 35: 2004 Sumatra (Chlieh et al, ; Subarya et al, ).…”