Crustal Deformation In The India-Eurasia Collision Zone From Long-term GPS Measurements

Studying crustal deformation is of great importance to understand the deformation mechanism of active structures and the dynamic process in the lithosphere,and assess seismic hazards.The India-Eurasia collision zone is the largest deforming region on the planet,the crustal deformation of which is a research hotspot of geoscience for a long time.GPS measurements can provide high-precision and large-scale quantitative information of the crustal deformation directly,thus it has been widely used to study the present-day crustal deformation.The higher the station density is and the longer the time span of the observation is,the more reliable the result is.Since the late 1980s,many projects have been carried out in the India-Eurasia collision zone for large-scale or regional GPS measurements,especially the CMONOC-?/? projects(Crustal Movement Observation Network of China/Tectonic and Environmental Observation Network of Mainland China),These projects have accumulated very abundant GPS data for the quantitative studies of the crustal deformation.This paper collects,processes and incorporates available GPS data as more as possible,and presents the GPS velocity field in the India-Eurasia collision zone with the highest spatial resolution and the longest time span so far.Based on the velocity field,this paper focuses on the deformation rates of the major active structures,strain rate distribution,crustal deformation mode,and seismic hazards in the India-Eurasia collision zone.The main work,results and conclusions are:(1)This paper uses the PANDA software developed at GNSS research center of Wuhan University to process the raw GPS data sampling at 30s of the CMONOC-I/II projects observed from 1998 when the projects started running to the end of 2015(i.e.,27 CMONOC-I continuous stations observed since 1999,56 CMONOC-I campaign stations observed annually during 1998-2007,1000 CMONOC-I campaign stations observed in 1999,2001,2004,2007,and 233 new continuous stations established by the CMONOC-II project operating since late 2010,2000 CMONOC-II campaign stations observed in 2009,2011,2013,2015),and obtains the time series of station positions with high-precision.Then correct the effects from the station events and 47 earthquakes during the time span of the GPS data.Subsequently,incorporate 1020 GPS stations outside of the CMONOC-?/?projects observed from 1991 to 2013.Finally,this paper obtains the interseismic GPS velocity field in the India-Eurasia collision zone including 2576 GPS stations observed from 1991 to 2015.(2)Based on the newest GPS velocity field,this paper updates the present-day deformation rates of the major active structures in the India-Eurasia collision zone.The updated results are:A)The convergence rates across the Tibetan plateau,Qaidam Basin-Qilian Shan,Liupan Shan,Longmen Shan and Himalaya are about 28-34,12,2,1-2 and 15 mm/yr,respectively;B)Along Tien Shan,the convergence rates decrease eastward from 19.0 to 2.2 mm/yr;C)The extensional rates are about 6,23,15 and 11 mm/yr in the northern,central,southern Tibetan plateau and Yunnan,respectively;D)In the eastern part of the India-Eurasia collision zone,the major strike-slip fault-the great Tanlu fault has a dextral rate of?1 mm/yr;E)In the western part of the India-Eurasia collision zone,the major Tibetan strike-slip faults are undergoing low strike-slip rates,mostly no more than 10 mm/yr.The sinistral rates across the western,central and eastern(near 95 °E)Altyn Tagh fault are 1.4mm/yr,8-9mm/yr and 4.5mm/yr,respectively.The main Haiyuan fault has a sinistral rate of-4-5 mm/yr.The sinistral rates across the Kunlun fault near 94°E,101°E and 103°E are 12.8,6.1 and 0.7 mm/yr,respectively.The Xianshuihe-Zemuhe fault has a sinistral rate of-9 mm/yr while the sinistral rate across the Xiaojiang fault is about 7 mm/yr.Across the Karakoram fault,the dextral rates near 77°E,79°E and 81°E are 6.0,3.8 and 3.2 mm/yr,respectively.The China segment of the Red River fault has a dextral rate of?3 mm/yr.(3)This paper studies some key issues for the major Tibetan strike-slip faults,including the locking depth,along-strike change in the slip rate and the comparision between geological and geodetic estimates of slip rates,which can be summarized as follows:A)Use the arctan function to calculate the locking depths of the faults.The results show that except for the western Altyn Tagh fault,the eastern Haiyuan fault and the China segment of the Red River fault,the locking depths for most of the faults are in the range of 13±6 km.The deep locking depths in the western Altyn Tagh fault and the China segment of the Red River fault suggest that other adjacent faults may be involved in these two locations,and the very shallow locking depth in the eastern Haiyuan fault is consistent with previous interpretations of shallow creep in this region;B)Discuss the changes in the slip rates along the strikes of the Altyn Tagh,Kunlun and Karakoram faults.The crustal shortening of the Qilian Shan thrust belt is responsible for the eastward-decreasing slip rate along the eastern Altyn Tagh fault.The similar change for the eastern segment of the Kunlun fault should be accommodated by the crustal thickening across the Anyemaqen Shan and the clockwise rotation of the eastern Kunlun fault.Along the Karakoram fault,the increasing dextral rates from southeast to northwest may be due to that the India-Eurasia convergence direction becomes more oblique to the Himalayan arc toward the west,resulting in more convergence is accommodated by the dextral slip of the Karakorum fault.In addition,this paper notes that variations in the slip rate along long faults may also be a useful test for models of along-strike fault propagation;C)Compare the new geodetic slip rates estimated by this paper against available geological estimates for the Quaternary to present or a shorter time span(e.g.,Holocene).The results show that for each fault,at least one geological estimate(usually the most recent/robust)is consistent with the geodetic rate.Also,this paper finds that the published geodetic and geological slip rates of the major Tibetan strike-slip faults gradually come together as time goes on.(4)Re-estimate the strain rate fields in the India-Eurasia collision zone using the newest GPS velocity field presented in this paper,including principal strain rate field,distribution of second invariant of strain rate tensor,maximum shear strain rate field and dilatational strain rate field.The new strain rate fields show four key features:A)There are several large regions with insignificant internal deformation,mainly including the Tarim-Gobi-Alashan region,and the Sichuan and Ordos basins;B)Strain concentration occurs around some of the major faults.The prominent amongstthese are the thrust Himalaya and Tien Shan orogenic belts,and the strike-slip Altyn Tagh,Kunlun,Xianshuihe-Xiaojiang,Sagaing faults,the strain rates of which are generally over 40 nanostrain/yr;C)There are regions of diffuse deformation.The most prominent is the Tibetan plateau,which is straining at an average rate of?18 nanostrain/yr away from several major faults with strain concentration;D)The high-elevation region of the Tibetan plateau is dilating significantly,and the dilation is strongest at elevations higher than?4750 m.The rates of horizontal dilation for the northern and southern plateau are similar at?7 and?5 nanostrain/yr,respectively.By comprehensive analyses and discussions,this paper finds that neither block nor continuum models in their purest forms can explain all the key features observed in the strain rate fields for the India-Eurasia collision zone.(5)Assess the global seismic hazard of the India-Eurasia collision zone based on the new strain rate field estimated by this paper.In about half of the study region,assuming that the typical seismogenic thickness is 15 km,the accumulation of elastic strain energy in the upper crust for a 100 km*100 km area can at least generate a Mw6.6 strong earthquake every 100 years.Discuss the seismic gaps in the main active faults in the India-Eurasia collision zone,including the central Altyn Tagh fault,the Tianzhu gap in the Haiyuan fault,the Xidatan-Dongdatan gap of the Kunlun fault,the seismic gaps along the Xianshuihe-Xiaojiang fault system,the western-central part of the Himalaya,and the central Sagaing fault.With the new strain rate field and the 1977-2016 GCMT catalog,this paper forecasts the shallow seismicity of the India-Eurasia collision zone,and infers that the region will suffer from about 11 Mw?7.5,35 Mw?7.0,107 Mw?6.5,and 320 Mw?6.0 earthquakes every 100 years.Compare the forecast using the 1977-1996 GCMT catalog with the actual situation of the 1997-2016 GCMT catalog.This paper finds that the forecast results for Mw7.0+earthquakes have a bias relative to the actual catalog which may be due to that the time span of 20 years is too short to include enough Mw7.0+ earthquakes in the catalog,while the forecast results for Mw6.0-7.0 earthquakes are close to the actual catalog,the success rates of which are generally over 80%,and nearly half are over 90%,implying a good quality for the forecast results.Describe two possible applications of the forecast result-identify seismic gaps with historical earthquake catalog,and distinguish between block and continuum models.