Crustal Deformation of the Altyn Tagh Fault Based on GPS

Abstract

Knowledge of the slip behavior of the Altyn Tagh fault (ATF) has significant implications for our understanding of the tectonic deformation of the Tibetan Plateau. In this study, we process Global Positioning System (GPS) data spanning 2009–2017 across the western ATF, merge the solution with recently published GPS velocities, and obtain a dense GPS velocity field for northern Tibet. We introduce an elastic block model and estimate the fault slip rate, interseismic fault coupling (ISC), and seismic moment accumulation rate along the ATF. The estimated left‐lateral strike‐slip rate of the ATF decreases eastward from 12.8 ± 0.4 mm/a to 0.1 ± 0.2 mm/a. Results show a heterogeneous distribution of ISC along the fault, with the fault locking depth varying from 5 to 20 km. The seismic moment accumulation rate is 2.16–2.37 × 1018 N m/km along three segments between Sulamu Tagh and Aksay bends, where the accumulated seismic moment over the last five centuries could be balanced by earthquakes with magnitudes of Mw 7.7, Mw 7.6, and Mw 7.8, respectively. We calculate the ratio of the Indo‐Eurasia convergence accommodated by the ATF and find that 10.3% of the convergence is accommodated along the fault by lateral extrusion; quantitative analysis of the strain rate, however, shows that a large part of the northern Tibet is not well described by elastic block rotation. Consequently, we suggest that a hybrid kinematic model that includes both block‐like and continuous deformation is needed to better delineate the crustal deformation of the Tibetan Plateau.

Geodetic Constraints on the Crustal Deformation along the Kunlun Fault and Its Tectonic Implications

“…Despite subtle differences in processing strategies and reference frames between the different solutions [41], the GPS site velocities represent the secular interseismic crustal deformation of the Tibetan Plateau before the 2001 Kokoxili earthquake. To combine them, we solved for the Euler vector between two solutions by minimizing the velocity residuals within common sites [22]. We choose the Zhang et al [32] solution as the master solution and translated the other velocity solutions into it.…”

“…During this time period, GPS sites were measured in 2015, 2016 and 2017, with at least 72 hours of continuous observation in each campaign. With reference to the GPS data processing strategies of Li et al [22], we used the GAMIT/GLOBK 10.6 software to process the raw GPS data [43]. We used IGS's precise orbits and Earth orientation parameters, absolute antenna phase center calibrations, and the FES2004 ocean tide-loading model to derive loosely constrained daily solutions, which were then expressed in the International Terrestrial Reference Frame (ITRF2008) using a standard 7-parameter transformation.…”

“…The fault coupling uncertainties are shown in Figure 6c-d; an overall uncertainty value of 0.3 was obtained. The ISC distribution, together with the estimated slip rate, was integrated to quantitatively calculate the seismic moment accumulation rate on the fault plane, and thus assess the seismic hazard on the Kunlun fault [22,37], which we also present in the discussion section.…”

“…Probing the crustal deformation of the Tibetan Plateau has long represented a significant research focus [3][4][5][6]. Over the last 30 years, modern geodetic observations (i.e., Global Positioning System (GPS) and interferometric synthetic aperture radar (InSAR)) provided abundant data and sufficient precision to investigate the large-scale surface deformation of the Tibetan Plateau [7][8][9][10][11][12][13][14][15]; and the interseismic, coseismic and postseismic faulting behaviors of large crustal faults in and around the Tibet, including the Main Himalaya Thrust [16,17], the Longmenshan fault system [9,18,19], the Xianshuihe-Xiaojiang fault system [20], and the Altyn Tagh-Haiyuan fault system [5,21,22]. However, relatively few studies have focused on the regions along the Kunlun fault (Figure 1), owing to insufficient coverage of GPS sites [7,9,11,12], which is the focus of this study.…”

Geodetic Model of the 2017 Mw 6.5 Mainling Earthquake Inferred from GPS and InSAR Data

“…The results show low b-values in the epicentral region prior to the Mainling earthquake, indicating a high stress state. We also calculated the strain rate of the eastern Himalayan syntaxis, based on the long-term GPS displacement rates [46]. Using an L-curve method [47], we interpolated the GPS velocity field by 0.4° × 0.4° and calculated the second invariant of the strain rate tensor (effective strain rate), which characterizes the strain state of the upper crust [47].…”