EGU General Assembly 2020
© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

Palaeo-earthquake magnitudes on the Dzhungarian fault, N. Tien shan, and implications for the rupture processes of intraplate strike-slip faults

Chia-Hsin Tsai1, Richard Walker1, Simon Daout1, Kanatbek Abdrakhmatov2, Aidyn Mukambayev3, Christoph Grützner4, and Ed Rhodes5
Chia-Hsin Tsai et al.
  • 1Department of Earth Sciences, University of Oxford, Oxford, United Kingdom of Great Britain and Northern Ireland (
  • 2Institute of Seismology, National Academy of Sciences, Bishkek, Kyrgyz Republic
  • 3Data Center of the Institute of Geophysical Researches, Almaty, Kazakhstan
  • 4Institute of Geological Sciences, Friedrich Schiller University Jena, Jena, Germany
  • 5Department of Geography, University of Sheffield, Sheffield, United Kingdom

Long-term and present-day crustal deformation in the northern Tien Shan is poorly known, but is a key to understanding the mode of lithospheric deformation deep within the continental interiors, as well as the hazards posed by the slow-moving intraplate faults. Driven by the India-Asia collision, the NW-SE strike-slip faults and the E-W range-front thrust faults in the interior of Tien Shan together accommodate about 15-20 mm/yr of shortening. Here we focus on the NW-SE striking Dzhungarian fault (DZF) and the E-W striking Lepsy fault (LPF), which are large oblique strike-slip faults bounding the Dzhungarian Alatau, northern Tien Shan. Two large historical earthquakes in ~1716 and 1812 (Mw 8) were recorded in this region, and clear fault traces as well as scarps are visible from satellite images along some of the main faults. However, their geometries, slip rates, mode of deformation, expected earthquake magnitudes and recurrence interval have not been studied in details. A previous study suggested that the LPF ruptured in a seismic event around 400 yrBP that might be the 1716 earthquake known from historical records. Offsets of over 15 m were found over a fault length of 120 km, indicating a magnitude in the range Mw 7.5-8.2. The slip to length ratio for the LPF is unusally high, suggesting either that faults in this region are capable of generating very large earthquakes for a given fault length, or that the rupture length is underestimated.

Using a combination of high-resolution digital elevation models (DEMs) and orthophotos from High Mountain Asia (NASA), Pleiades optical imagery (CNES), drone photos and multi-temporal interferometric synthetic-aperture radar (InSAR) from the Sentinel-1 satellites, we identify the geomorphic signatures and quantify the long-term and short-term strain accumulation along the faults. The ~400 km DZF shows evidence for relatively ‘fresh’ rupturing along much of its length. We calculate an average lateral slip per event of 9.9 m from offset stacking analysis, which underlines the potential future large earthquakes on this fault. The proximity of the DZF and LPF ruptures and equivalent level of preservation opens the possibility that they were formed in a single earthquake event, with a moment-magnitude greater than 8. We also present estimates of long-term and short-term rates of slip across the DZF in order to estimate average recurrence intervals and to build a kinematic model of the faulting in the Northern Tien Shan.

How to cite: Tsai, C.-H., Walker, R., Daout, S., Abdrakhmatov, K., Mukambayev, A., Grützner, C., and Rhodes, E.: Palaeo-earthquake magnitudes on the Dzhungarian fault, N. Tien shan, and implications for the rupture processes of intraplate strike-slip faults, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-710,, 2019

Comments on the presentation

AC: Author Comment | CC: Community Comment | Report abuse

Presentation version 2 – uploaded on 05 May 2020
Slip-rate section and more figures are put in the poster.
  • CC1: Comment on EGU2020-710, Dan Clark, 07 May 2020


    Could you please explain the "offset stacking analysis" process that you used? Also, by how much might the results of the analysis locally under-estimate slip due to the vertical component to motion not being included (e.g. in section S2)?


    Dan Clark (Geosceince Australia)

    • AC1: Reply to CC1, Chia-Hsin Tsai, 07 May 2020

      Hi Dan,

          Thank you for bringing up these questions.
          The offset stacking analysis mentioned in the abstract is similar to the one used in this paper:
      Kurtz, R., Klinger, Y., Ferry, M., & Ritz, J. F. (2018). Horizontal surface-slip distribution through several seismic cycles: The Eastern Bogd fault, Gobi-Altai, Mongolia. Tectonophysics, 734–735(March), 167–182.

      However, in the poster, I decided not to demonstrate and use the result from the stacking analysis. Because we found some possible biases in this method and we'd like to adjust it a bit to get a more convincing result from the stacking analysis. So the amount of average lateral slip in the southern DZF in the poster is defined by the actual measurements along the fault only but not the stacking result. (But it seems the stacking result would be around 9 m anyway, which is quite similar to the actual measurements.)  

      As for the second question, the vertical component in the northern DZF (S2) is actually included for the estimation of magnitudes. We deduce the vertical component on the northern fault plane by assuming 1) the same lateral slip vector as the south, 2) a dipping angle of 60 degrees, and 3) the ratio of lateral to vertical offset found from S2. The fig. 7 in the poster is explaining this. Nevertheless, a huge part of the uncertainty would be on the dipping angle since we don't have any clue for it.

      I hope this will answer your questions.

      Many thanks,


Presentation version 1 – uploaded on 05 May 2020 , no comments