This study investigates the intricate relationship between earthquake sources and seismogenic surface ruptures in a complex tectonic setting with active faults in the continental collision zone between the southern Tien Shan and the northern Pamir Mountains in Central Asia. The study focuses on the 2008 M_w 6.6 Nura earthquake along the Pamir Frontal Thrust, where the seismogenic surface rupture occurred unexpectedly within the footwall and 10 km away from the source thrust fault. This discrepancy raises questions about the interactions and potential trigger mechanisms between tectonic structures during earthquake rupture. Using unmanned aerial vehicle photography and field inspection, our investigation integrates detailed fault-zone mapping with tectono-geomorphic observations to unravel potential interactions between subsurface structures and surface-deformation phenomena. Our findings suggest that a combination of slip along deep-seated basement faults and remotely triggered flexural slip within folded Paleogene strata led to surface rupture of overlying Quaternary glacial deposits. Geomorphological and geochronological analyses coupled with systematic displacement measurements furthermore reveal evidence of similar past ruptures within the regional fault system, suggesting a recurrence interval of 1.7 kyr and a Holocene vertical offset rate of 0.4 mm/yr. The analysis of the Nura rupture zone contributes significantly to evaluate linkages between surface and subsurface structures regarding fault-zone behavior and seismic hazard assessments. Importantly, our results highlight the critical role of on-site investigations in regions with poorly defined surface ruptures, where misinterpretation may lead to the underestimation of the impact of seismic events and limitations in assessing earthquake history and strain accumulation.

Key Points

Field survey of the 2008 M_w 6.6 Nura earthquake surface rupture in Kyrgyz Pamir shows evidence of secondary flexural-slip faulting
Our results suggest similar events in the eastern Pamir-Tien Shan collision zone with a recurrence interval of approximately 1.7 kyr
We propose a rupture model where slip along deep-seated tectonic structures remotely triggered flexural-slip along shallow folds

Plain Language Summary

This study investigates surface ruptures related to earthquakes in Central Asia, focusing on the 2008 Nura earthquake in south Kyrgyzstan, where the Tien Shan and Pamir mountains converge due to the collision between the Indian and Eurasian tectonic plates. The surface rupture during that event occurred at a different location than expected, causing us to investigate more closely the complex interactions between faults during earthquakes in this region. Based on detailed mapping on a high-resolution digital surface model and on-site observations, our analysis suggests that movement along hidden tectonic fault structures in the subsurface caused the bending of overlying folded sedimentary strata, which in turn resulted in the breaking of the ground. Through geomorphological analysis and geological age determination, we conclude that similar events in the regional fault system occurred in the past, in intervals of 1,700 years and with a vertical growth rate of 0.4 mm/yr within the last 12,000 years. Understanding such geotectonic processes is important for a better assessment not only of earthquake hazard in this region, but also of faulting processes elsewhere that have a similar deformation character, thus emphasizing the importance of field investigations in regions where the understanding of tectonic structures is limited.

1 Introduction

Due to complex spatiotemporal structural characteristics and lithological juxtapositions, the investigation of destructive earthquakes in continental collision zones is a challenge for earthquake geology research and seismogenic hazard evaluations (e.g., Bilham et al., 2001; Elliott et al., 2016; Molnar & Ghose, 2000). Assessing the hazard potential of faults is especially difficult when seismic events result in subtle surface manifestations of deformation (e.g., 1974 M_w 7.1 Markansu earthquake, Kyrgyzstan; Fan et al., 1994; Jackson et al., 1979; Nikonov et al., 1983) or when the relationship between earthquake foci, seismogenic rupture, and deformation is enigmatic (e.g., 1944 M_w 7 San Juan earthquake, Argentina; Rockwell et al., 2014). The latter case poses a major obstacle in assessing the seismogenic character of neotectonic faults because coseismic surface ruptures during a destructive earthquake may correspond to the surface expression of second-order faults. Such second-order structures may be flexural-slip or bending-moment faults, closely related either to buried thrusts beneath the ruptured areas (e.g., Philip & Meghraoui, 1983; Yeats, 1986) or to hard-linked faults located farther from the rupture that trigger deformation in adjacent, mechanically weak strata (e.g., Kaneko et al., 2015). The ambiguity surrounding these relationships not only risks underestimating the potential impact of seismic events but also hampers our understanding of earthquake history and strain accumulation over time. Unraveling such complexities is crucial, as the seismogenic potential of these regions often remains unrecognized until a destructive earthquake occurs, as exemplified by the 1988 M_w 6.8 Spitak earthquake in Armenia (Philip et al., 1992) and the 1994 M_w 6.7 Northridge earthquake in California (Hauksson et al., 1995). To gain a better understanding of the interaction between deep-seated, hidden structures and their surface expression during rupture, as well as the potential transfer of motion between spatially separate structures, a detailed geological and tectono-geomorphic field analysis, combined with geodetic and geophysical investigations, is necessary. To address these critical issues, we examined a fault rupture associated with the 2008 M_w 6.6 Nura earthquake in the northern Pamir Mountains of Central Asia within the continental collision zone between India and Eurasia, specifically focusing on the complex tectonic setting of the region and its influence on the spatial pattern of surface ruptures.

The Northern Pamir, with elevation peaks as high as 7 km and separated from the neighboring Tien Shan mountain belt by the intermontane Alai Basin to the north, has a long history of deformation that occurred in both pre-Cenozoic and Cenozoic mountain-building (e.g., Burtman & Molnar, 1993; Sobel et al., 2013; Strecker et al., 1995). The late Cenozoic landscape of the Northern Pamir bears the traces of climate-controlled surface processes, including remnants of multiple valley and piedmont glaciations (e.g., Nöth, 1932; Strecker et al., 2003), fluvial terrace formation (Arrowsmith & Strecker, 1999; Nikonov et al., 1983; Patyniak et al., 2021), mass-movement deposits and associated valley impoundments (Reznichenko et al., 2017; Robinson et al., 2015), as well as superposed faults and tectono-geomorphic phenomena. The collisional tectonic processes have resulted in a series of faulted and folded rock formations and unconsolidated deposits, reactivated crustal heterogeneities, and kinematically linked Quaternary fault arrays that record the ongoing northward motion of the Pamir (e.g., Burtman & Molnar, 1993; Coutand et al., 2002; Nikonov et al., 1983; Nöth, 1932; Patyniak et al., 2021; Strecker et al., 1995). The erosional power of the westward-flowing Kyzilsu River, the main drainage system of the Alai Basin, ensures the removal of material delivered by tectonics along the Pamir Thrust System (PTS) (Figure 1), maintaining a stable mountain front in the western and central Alai Basin. However, in the Nura region to the east in the upper part of the watershed, insufficient erosional capacity leads to material accumulation, resulting in out-of-sequence faulting and an indistinct range front (Pavlis et al., 1997; Strecker et al., 2003).

Details are in the caption following the image — **Figure 1**
Open in figure viewer PowerPoint

Seismotectonic overview of the Northern Pamir and Southwest Tien Shan collision zone. (a) Overview of recent seismicity of the eastern part of the Northern Pamir range front, the Alai Basin and the Pamir-Tien Shan collision zone with major fault structures and earthquake locations. Shaded area represents the Pamir Thrust System. Colored circles indicate main shock and aftershock sequences related to the 2008 Nura earthquake from a regional network between 2008 and 2014 (Sippl et al., 2014). Black circles represent seismicity from the USGS earthquake catalog. Focal mechanisms are from the Global Centroid-Moment-Tensor catalog (Dziewonski et al., 1981; Ekström et al., 2012), Fan et al. (1994) and Sippl et al. (2014). The isoseismal contours are from USGS ShakeMap. Base map hillshade is a 30-m resolution NASA Shuttle Radar Topography Mission digital elevation model. DF = Darvaz Fault, KES = Kongur Extensional System, KYS = IrkF = Irkeshtam Fault, PTS = Pamir Thrust System, PFT = Pamir Frontal Thrust, MPT = Main Pamir Thrust, STST = South Tien Shan Thrust. (b) Simplified regional tectonic setting and location of study area (shaded rectangle) in Central Asia where the northward translation of the Pamir indenter causes deformation in the transition with the Tien Shan. Black arrow indicates global navigation system horizontal surface velocity relative to stable Eurasia (e.g., Zubovich et al., 2010).

Seismicity in the Northern Pamir is concentrated along the PTS (Fan et al., 1994; Schurr et al., 2014), which serves as the primary fault system that accommodates northward propagation of the Pamir (Arrowsmith & Strecker, 1999; Burtman & Molnar, 1993; Coutand et al., 2002; Patyniak et al., 2021; Strecker et al., 2003; J. A.Thompson et al., 2015). In recent decades, the eastern continuation of the PTS has experienced several earthquakes with a magnitude greater than M_w 6.5, most notably the M_w 7.1 Markansu event in 1974 (Fan et al., 1994). However, despite the well-documented Quaternary faults and associated deformation structures, the characterization of recent and historical earthquakes in the eastern part of the PTS is difficult due to the wide distribution of deformation phenomena and the lack of unambiguous surface ruptures (Fan et al., 1994; Jackson et al., 1979; Nikonov et al., 1983). In 2008, the M_w 6.6 Nura earthquake struck along the Pamir Frontal Thrust (PFT) at the leading edge of the eastern PTS (Figure 1), triggered a series of aftershocks and completely destroyed the Nura settlement, resulting in the loss of 74 lives and considerable damage to infrastructure (Abdrakhmatov et al., 2008; Kalmetieva et al., 2009; Qiao et al., 2015; Sippl et al., 2014; Teshebaeva et al., 2014). Interestingly, inversion of InSAR data that recorded this event shows that the earthquake was associated with a surface rupture that broke in the footwall of the PTS, 10 km away from the source fault (Teshebaeva et al., 2014). Subsequent seismological investigations revealed that the hypocenter of this event was at ∼3.5 km depth beneath the Trans Alai Range (Sippl et al., 2014). That no surface deformation was observed along the bounding fault of the Trans Alai Range suggests that strain and slip partitioning as well as diffuse deformation across the principal regional tectonic structures must have played a significant role in the rupture process (Qiao et al., 2015; Sippl et al., 2014; Teshebaeva et al., 2014).

In this study, we build on these InSAR and seismological studies to present our field-based analysis of a seismogenic surface rupture associated with the 2008 M_w 6.6 Nura earthquake that occurred in a geologically complex area beyond the source fault. We examined the surface rupture caused by the earthquake by integrating tectono-geomorphic observations with detailed fault-zone mapping on high-resolution digital surface models (DSM) and newly acquired geochronological data. Our structural data provide valuable insight into faulting mechanics and suggests the nature of the subsurface structures responsible for surface deformation observed.

2 Background

The 2008 M_w 6.6 Nura earthquake occurred on October 5 at 15:52 (Greenwich Time) south of the Nura settlement in southern Kyrgyzstan (Figure 1). Seismic shaking affected distal areas in the north, including the provinces of Osh, Batken, Jalal-Abad, and Naryn in Kyrgyzstan, and the border areas of Uzbekistan, Tajikistan, and China (Abdrakhmatov et al., 2008; Kalmetieva et al., 2009). Despite the magnitude of the main shock and its shallow depth of 3.5 km (Sippl et al., 2014), no surface rupture was reported along the structures defining the PFT, which are inferred to be linked to the hypocenter. Instead, the surface broke about 10 km north of the hypocenter within the footwall of the PFT through portions of the upper slopes of the northwestern riverbank and upslope area of the Nura River valley, defined as the Irkeshtam fault (IrkF) (Qiao et al., 2015; Sippl et al., 2014; Teshebaeva et al., 2014). This zone is located within a structural transition of the collision zone between the Trans Alai Range of the Northern Pamir and the Alai Range of the southern Tien Shan (Figure 1).

2.1 Seismological Characteristics

A point source deviatoric moment tensor determined from regional stations (i.e., Sippl et al., 2014) and a Global Centroid-Moment-Tensor (CMT) solution (http://globalcmt.org) record an almost pure reverse-faulting mechanism for the 2008 Nura main shock, with an approximately east-west oriented strike and 59° south-dipping fault plane (Figure 1). The main event was followed by a series of aftershocks with hypocentral maximum depths of 17 km, including the two strongest aftershocks; one of these events had an estimated magnitude of M_w 5.4 that ruptured close to and perhaps along the same fault as the main shock (Sippl et al., 2014). Based on the time of initiation and swarm distribution with respect to the regional faults, Sippl et al. (2014) divided the aftershocks into five clusters (Figure 1), which are summarized below.

The main event was associated with a distinct east-west trending cluster to its south and with a clear relationship to the PFT (red cluster on Figure 1). Approximately 20 km to the northwest, near the eastern terminus of the Alai Basin, a small cluster dominated by strike-slip faulting along northeast- or northwest-striking fault planes was triggered coevally with the main shock (purple cluster). Shortly after the main event, another cluster of aftershocks initiated with reverse-fault mechanisms and subvertical, east-west-striking rupture planes (orange cluster). Roughly 80 min after the main shock, a distinct cluster of aftershocks (blue cluster) occurred at the eastern end of the main shock cluster (red cluster). Seismic activity in the blue cluster shifted toward the north-northeast and north of the PFT. About 20 days after the main event, a fifth cluster of aftershocks (yellow cluster) initiated east of the main shock and the blue cluster. The earthquakes with thrust and reverse mechanisms within the yellow cluster were primarily oriented east-northeast following the local topography and the slight deviation of the PFT toward the northeast (Sippl et al., 2014).

Sippl et al. (2014) pointed out that the blue cluster aligned subparallel with the IrkF and followed the northeast-oriented structural trend of the Tien Shan located to its north, and that the events crossed the topographic and structural trends of the imbricate PTS faults (Figure 1). The authors therefore suggested that the trace of the IrkF could potentially indicate the leading edge of a steeply southeast-dipping fault associated with a ramp anticline. The blue cluster was interpreted to reflect activity of the deeper segments of this ramp, possibly corresponding to an inherited, northeast-striking crustal heterogeneity within the Tian Shan. Consequently, the main shock of the 2008 Nura event may have activated the steeply south-dipping PFT, which trends approximately eastward; the subsequent increase in Coulomb stresses in the surrounding volume subsequently triggered the aftershock series within the deeper structure of the Paleozoic to Miocene orogenic wedge north of the thrust (Sippl et al., 2014).

2.2 Regional Tectonics and Geology

The Trans Alai Range is bounded by north-verging thrust faults of the PTS (Figure 1) (Arrowsmith & Strecker, 1999; Li et al., 2019; Nikonov et al., 1983; Nöeth, 1932; Sobel et al., 2013; Strecker et al., 2003; J. A. Thompson et al., 2015). The PTS is interpreted as the up-dip expression of a continental subduction zone with an orogenic accretionary wedge (Burtman & Molnar, 1993; Hamburger et al., 1992; Sobel et al., 2013). The PTS comprises south-dipping thrust faults and can be divided from south (old) to north (young) into the Main Pamir Thrust (also defined as the Markansu Fault in other studies; e.g., Coutand et al., 2002; Qiao et al., 2017; Strecker et al., 1995, 2003; Xiong et al., 2019) and multiple splays of the PFT (Arrowsmith & Strecker, 1999; Coutand et al., 2002; Sobel et al., 2013; Strecker et al., 1995, 2003). East of ∼73°20'E, the PTS widens northward and merges with the south-vergent fold-and-thrust structures of the southern Tien Shan, separating the once contiguous Alai and Tarim basins (see Figure 1) (e.g., Burtman & Molnar, 1993; Ge et al., 2022; J. A. Thompson et al., 2015; Thompson Jobe et al., 2017).

In the area affected by the Nura earthquake, the imbricated PFT faults successively involve a sequence of steeply dipping Cretaceous to Paleogene sedimentary strata. This sequence occurs repeatedly, but spatially limited, in the southern parts of the Tien Shan and in the Tadjik fold-and-thrust belt (e.g., Chapman et al., 2017; Pickering et al., 2008). At approximately 73°55'E and north of the PFT near the Nura settlement, a comparable sequence of northeast-trending folded Cretaceous to Paleogene strata is bounded by folded Silurian-Devonian units and partially covered by northwest-dipping Neogene conglomerates (Figure 2). The northwest-dipping syncline limbs are repeatedly exposed along strike and extend for about 8 km northeast of the Nura settlement, as visible on Google Earth satellite imagery (Figure 2). Interestingly, southwest of the Nura settlement, subparallel to the trend of these synclinal limbs and along the northwestern bank of the Nura River, runs the trace of the IrkF, defined as a left-lateral transpressional fault that accommodated the rupture associated with the 2008 Nura earthquake (Teshebaeva et al., 2014). From there, the IrkF extends farther toward the southwest and appears to eventually connect obliquely with the PFT (Figures 1 and 2).

To the west of the Nura River valley, extensive areas are covered by different generations of glacial till whose corresponding glaciers were sourced in the Trans Alai Range (Figure 2). These deposits delimit the eastern terminus of the intermontane Alai Basin. Based on their differing overall geomorphic expressions of associated landforms, they have been subdivided into four separate generations (Qm1–4; old to young) corresponding to the sequence found in the Alai region as defined by Arrowsmith and Strecker (1999) and Patyniak et al. (2021), using the nomenclature proposed by Nikonov et al. (1983). The area affected by the IrkF, where the surface was ruptured during the Nura earthquake, is covered by Qm2 tills that abut Qm1 deposits away from the Nura River. In accordance with stratigraphic and geochronological relationships in the Alai Basin, the Qm2 glaciation was demonstrably an early Holocene event (Arrowsmith & Strecker, 1999; Patyniak et al., 2021; Strecker et al., 2003). The former glacier responsible for the Qm2 deposits appears to have incised the Qm1 moraines in a north-deflected direction near the Nura settlement, virtually along the contact with the folded Cretaceous to Paleogene sequence, creating a 100-m-deep valley (at ∼39°39'N 73°50'E in Figure 2 and video available at Zenodo; see Data Availability Statement Section). Within the Qm2 deposits on the northwestern slopes of the Nura River valley and along the Nura earthquake surface rupture, Paleogene gypsum-bearing strata are exposed.

Several streams that originate in the Trans Alai Range (Northern Pamir) in the south and the Alai Range (Tien Shan) in the north join the eastward-flowing Kyzilsu River in the vicinity of the rupture zone (Figure 2; see also discussion in Strecker et al., 2003). Note that this river has the same name as the westward-flowing Kyzilsu River in the Alai Basin farther west. The Nura River is sourced from and flows northward through the imbricate PFT folds and thrusts of the Trans Alai Range. Immediately south of the IrkF, at ∼39°35'N, the river changes course to the northeast and continues to flow parallel to the fault and the exposed synclinal limbs described above, eventually joining the eastward-flowing Kyzilsu River. The course of the Nura River follows a zone that separates the Paleozoic rocks from Paleogene sedimentary strata and overlying Quaternary glaciofluvial deposits.

2.3 Geodetic and Structural Observations Across the Irkeshtam Fault Zone

Prior to our analysis, two InSAR studies were conducted in the region to quantify the deformation patterns and changes in surface topography resulting from the 2008 Nura seismic sequence (Qiao et al., 2017; Teshebaeva et al., 2014). Due to the local constraints of high elevation and snow cover, the InSAR measurements were not well resolved along the PFT. Nevertheless, north of the PFT along the IrkF, a distinct signal revealed surface deformation for which no seismicity data had been recorded. The InSAR-modeled surface deformation extends northeast-southwest for ∼25 km and follows the strike of the IrkF. The fault model by Teshebaeva et al. (2014) defines the IrkF as a south-dipping fault with a change in dip of 37°–44° from east to west, respectively. An oblique thrust-fault mechanism with ∼2 m of vertical offset and ∼1 m of left-lateral strike-slip offset were inferred. By incorporating regional geological observations, Teshebaeva et al. (2014) proposed that the IrkF is rooted in a shared décollement with the PFT. Furthermore, they argued that the IrkF is an integral part of a deformation zone where predominantly aseismic motion occurs within mechanically weak strata enabled by the gypsum-bearing Paleogene strata. Teshebaeva et al. (2014) further reported a prominent 7-km-long surface rupture along the IrkF southwest of the Nura settlement. This surface rupture, which is the main subject of this study, produced steep scarps, with vertical offsets of up to ∼80 cm and evidence of left-lateral motion, consistent with the sense of motion along the deformation zone determined by their analysis of InSAR-measurements.

3 Methods

3.1 Field Observations

During our field campaign in 2018, an extensive displacement-vector survey of the surface breaks was performed and cataloged to create a systematic offset documentation. The survey was subdivided into four 500- to 1,000-m-long blocks 1–4 (Figure S1 in Supporting Information S1). Displacement-vector measurements were taken every 10 m in each block resulting in 119 field measurements. These include components of fault slip and sense of displacement derived from local trace bearing (strike), vertical displacement, opening distance, and opening direction (azimuth of the slip vector) when a clear piercing point was available. To record the exact locations of the ground observations for cartography and to trace selected surface breaks, we used a kinematic differential global navigation system (dGNSS). Field measurements and associated imagery are available at Zenodo (see Data Availability Statement Section).

3.2 Infrared Stimulated Optical Luminescence

We obtained samples of predominantly silty sand units for infrared stimulated luminescence dating (IRSL) using aluminum tubes with a diameter of 5 cm (Figure S2 in Supporting Information S1). The tubes were hammered horizontally into cleaned vertical stratigraphic sections. These samples were retrieved and securely sealed to prevent exposure to light. Heterogeneous units with coarser grain sizes were sampled manually at night using a low-intensity red LED headlamp and placed in double-bagged 4 mil black conductive bags. In addition, we collected material from the surrounding area to estimate the dose rate through low-level gamma-spectrometry. All samples were processed at the Institute of Earth and Environmental Science, University of Freiburg, Germany. Detailed information on the IRSL processing can be found in Text S1 in Supporting Information S1.

Similar to previous studies conducted in active mountain belts (Preusser et al., 2006), the investigated quartz grains displayed low luminescence sensitivity. Therefore, feldspars were measured using the single aliquot regenerative dose protocol developed by Reimann and Tsukamoto (2012). During storage tests, minimal or no signal loss (<1 g/decade) was observed, indicating the absence of fading in the samples examined. The summarized results can be found in Table 1. It is worth noting that for both IRSL and post-IR IRSL (pIR), we noticed minor discrepancies in the ages determined by the two approaches. The pIR ages were consistently slightly but not systematically older than the IRSL ages (offset between 16% and 40%). We interpret this as an indication of partial bleaching of the pIR signal (as discussed by Gray et al., 2015), rather than an effect of fading, as the latter would likely result in a systematic offset. As a result, we exclusively base our interpretations on the IRSL ages.

Table 1. Summary of Infrared Stimulated Luminescence (IRSL) and Post-IR IRSL (IRSL at 150°C) Ages

Sample name	Depth (cm)	W (%)	K (%)	Th (ppm)	U (ppm)	D (Gy/ka)	Grain size (μm)	OD	Model	De IRSL (Gy)	De pIR-150 (Gy)	Age IRSL (ka)	Age pIR-150 (ka)
NU18-O1	75	15 ± 5	1.93 ± 0.13	9.88 ± 0.64	3.08 ± 0.19	3.74 ± 0.17	100–150	0.15/0.29	MAM/MAM	12.78 ± 0.32	15.63 ± 0.22	3.42 ± 0.16	4.19 ± 0.17
NU18-O2	225	15 ± 5	1.69 ± 0.14	9.26 ± 0.60	2.90 ± 0.18	3.39 ± 0.14	100–150	0.03/0.02	CAM/CAM	38.49 ± 0.25	53.75 ± 0.46	11.3 ± 0.5	15.8 ± 0.7
NU18-LO1	383	15 ± 5	1.56 ± 0.13	6.21 ± 0.41	1.93 ± 0.13	2.81 ± 0.13	100–150	0.06/0.06	CAM/CAM	33.68 ± 0.38	46.91 ± 0.54	12.0 ± 0.6	16.7 ± 0.8
NU18-TO1	270	15 ± 5	2.46 ± 0.25	7.22 ± 0.46	1.96 ± 0.12	3.91 ± 0.22	200–250	0.07/0.11	CAM/MAM	5.44 ± 0.06	6.29 ± 0.09	1.39 ± 0.08	1.61 ± 0.09

Note. Displayed are the assumed average water content during burial (W), the concentration of dose-rate relevant elements (K, Th, and U), dose rate (D), observed overdispersion (OD), and equivalent dose (De). A total of 40 aliquots were measured per samples, the vast majority of which (>95%) passed the usual quality criteria (recycling ratio within 10%, recuperation <10%, test dose >3* background signal). Water content was estimated based on present day moisture and grain-size composition. Mean equivalent dose was calculated using either the Central Age Model (CAM) or the Minimum Age Model (MAM), depending on the statistical parameters of De distributions, such as overdispersion, skewness and kurtosis (cf. Galbraith & Roberts, 2012). Ages were calculated using ADELEv.2017 (Degering & Degering, 2020), using the results of high-resolution gamma spectrometry (Preusser et al., 2023).

3.3 Digital Surface Models

DSM were generated from low-altitude aerial photographs taken at altitudes between 35 and 120 m above the launch point using an unmanned aerial vehicle (UAV); see Text S2 Supporting Information S1 for technical details. The UAV flight survey of ∼8 km along strike of the 2008 Nura surface rupture covered an area of approximately ∼5.5 km². In total, 14,156 photographs and videos were acquired, and 27 ground-control points (GCPs) were georeferenced with dGNSS. We used the photogrammetric modeling software Agisoft PhotoScan Pro (v.1.2.1 and v.1.4.3) to generate high-resolution point clouds and DSMs (available at OpenTopography; see Data Availability Statement Section) and perform image-based modeling by feature recognition and photo alignment using the Structure-from-Motion (SfM) algorithm (e.g., Bemis et al., 2014; Johnson et al., 2014). In total, this survey yielded 11 overlapping dense point clouds, orthomosaics, and DSM with a resolution of 4–10 cm/px, and a georeferencing constraint of ∼1–2 GCP/km² (Table S1 in Supporting Information S1). Three of the DSMs (set 9, 10, and 11) resulted in poorly georeferenced models due to insufficient GCP distribution and were excluded from further analysis (Figure S3 in Supporting Information S1). For simplicity, throughout the following text we refer to locations within the DSM by their distance from the northeastern end of the DSM, the point of origin, starting from kilometer 0.

The DSM data and the computed topographic derivative maps were used to map surface-rupture traces in ArcMap® at a fixed scale of 1:400. During the digital surface-break mapping, several attributes were assigned to the rupture traces: an identification number; scarp-face direction; start/end point; length; strike and type (principal displacement zone (PDZ) or distributed). The “type” was assigned on the basis of trace continuity, length, and/or vertical separation (VS) within the areas where the fault zone consists of multiple surface rupture splays. During fault-trace mapping, surface breaks visible in the discarded DSMs (set 9, 10, and 11) were mapped and subsequently moved to the correct location relative to the overlapping georeferenced map. While this portion of the mapping will be considered for general discussion of the rupture extent and distribution, it is excluded from all further calculations. The shapefile is available at Zenodo (see Data Availability Statement Section).

3.4 Vertical Separation Measurements

Topographic fault-scarp profiles oriented perpendicular to the fault scarp were extracted from the DSMs generated in this study to estimate the VS of single-event surface breaks and associated topographic highs, respectively. Here, the VS should not be confused with the fault throw (vertical fault slip), which requires knowledge of the near-surface fault dip and accounts for non-parallel far-field surface slopes (e.g., Johnson et al., 2018; S. C. Thompson et al., 2002). For VS measurement along the surface rupture, profile locations were systematically generated every 50 m along strike to avoid user-induced bias. In some areas along the surface rupture the offsets are small (minimum of several decimeters) and distributed across complex surface rupture patterns (i.e., multiple fault splays along one profile). In such cases modeled VS along each offset within one profile were summed up into net slip.

We employed a Matlab script developed by DuRoss et al. (2019) to perform the VS modeling. This script utilizes a linear approximation method to determine the footwall, hanging wall, and scarp face and project the geomorphic surfaces to the midpoint of the scarp (see also Bello et al., 2021). This projection enables the modeling of elevation differences between the offset surfaces, measured at the midpoint of the scarp. The precision of the modeled VS is mainly limited by the quality and resolution of the DSM. During linear regression, uncertainties in modeled slope angles arise from (a) natural undulations in the ground surface, and/or DSM artifacts, and/or presence of vegetation, and (b) unquantifiable “user error” in the subjective selection of profile sections during the algorithm execution. These uncertainties in turn affect the calculation of the VS. For error reduction, the script allows multiple VS-measurements along one profile and multiple iterations for each individual offset, and provides minimum and maximum VS-values (including the uncertainties described above).

4 Results

4.1 Morphological Observations and Integrated Chronostratigraphic Insights

We investigated the fault-zone geology and geomorphology of the IrkF at several locations along the A371 road, which runs from the Alai Basin through the Nura settlement across the border with China (Figure 3). The road was built in 2005 and repaired in 2015. In the Kyrgyz national earthquake reviews following the 2008 earthquake, Abdrakhmatov et al. (2008) and Kalmetieva et al. (2009) reported ruptures in the asphalt in the section of the road that passes through the Nura settlement and over the Nura River bridge, which were subparallel to and along the strike of the surface rupture. On the western side of the bridge, we found repaired sections of the road where rupture segments parallel to the trace of the IrkF cut through the Qm2 moraine deposits (Figure 3). Here, these deposits are characterized by subangular to subrounded boulders up to several meters in size in a coarse matrix and are covered by a layer of reworked loess containing rock fragments. A shear zone in these strata provides information about the orientation and dip of the potential fault plane, although with limited outcrop. The exposed till is at least 10 m thick, but the thickness of the re-transported loess layer varies laterally from tens of centimeters to several meters. To the southwest, a >2-m-thick reworked loess layer was identified in an abandoned military trench. IRSL dating (NU18-O1 and NU18-O2; Table 1) yields ages of 11.3 ± 0.5 and 3.4 ± 0.2 ka, from bottom to top (Figure 3; see also Figure S2 in Supporting Information S1).

The Nura settlement is located in the hanging wall of the IrkF and was built on a fluvial terrace adjacent to lateral moraine deposits (Qm3 deposits). A planar stratal surface is exposed northeast of the settlement that defines the southern limb of a syncline in Paleogene rocks. Here, a series of mass-movement deposits that originated from failure of rocks in the folded topography cover a terrace of the Nura River (Figure 3). The surface roughness of these deposits varies, allowing subdivision into three different mass-movement generations (L1–L3, from old to young). The mass-movement deposits are cut by the A371 road; in the cross section along the road, beds of lacustrine silty fine sand abut coarse mass-movement deposits at the edge of L1, suggesting transitory lake formation along the Nura River during landslide-related (L1) valley impoundment. IRSL dating of the paleo-lake deposits yields an age of 12.0 ± 0.6 ka (NU18-LO1; Table 1). Farther to the southwest, the contact between mass-movement deposit L3 and the Nura terrace is exposed along the river cut. An IRSL sample was collected from strata beneath the contact point with mass-movement deposit L3. The result of the analysis yields an age of 1.4 ± 0.1 ka (NU18-T1; Table 1).

Farther north, near the confluence of the Nura and Kyzilsu rivers, northwest-dipping Neogene sediments are covered by another mass-movement deposit on the opposite side of the Irkeshtam settlement, northeast of Nura (Figure S4 in Supporting Information S1). Field observations combined with surface geomorphology from ArcMap World Imagery Basemap and TanDEM-X images suggest that the northern mass-movement deposits correspond to a single event of unknown age. It is worth noting that extensive areas north of the Kaltabulak River that are mapped as Jurassic in the Russian maps were designated as Paleogene by Teshebaeva et al. (2014) and played a crucial role in their discussion on fault models. However, considering the stratigraphic characteristics of these sediments (Figure S4 in Supporting Information S1), it is more likely that these units constitute rock-avalanche deposits involving Jurassic rocks.

Along the rupture of the 2008 Nura earthquake itself, we observed numerous secondary geomorphic features associated with fault-zone deformation and subsequent erosional overprint, including fault-line sinkholes, ponded sediments on the footwall of the fault, and different types of mass-movement deposits (Figure 4). Most of the latter are topographically below the 2008 rupture in the high slope areas of the hanging wall of the IrkF, with an increased occurrence at the southwestern end and beyond the rupture zone. These mass-movement deposits are large enough to be recognized on Google Maps® satellite imagery, allowing comparison between pre- and post-2008 images. With the exception of one landslide deposit documented in 2010, the rest of these deposits predate the 2008 Nura earthquake (Figure S5 in Supporting Information S1).

Several locations along the rupture zone exhibit an increase in vegetation density, indicating enhanced water availability due to the formation of sag ponds against the scarp (Figure 4). Within the southwestern terminus of the fault zone (kilometers 5 to 6; Figure 4), an area of approximately 0.1 square kilometers has significantly denser vegetation cover and greater water availability than the surrounding regions within the fault zone. Notably, numerous fault-line sinkholes and a spring adjacent to the 2008 rupture are observed in this sector of the rupture zone. The stream channel network along the Nura River flank features relatively straight, closely spaced (<100 m) and barely branched northwest-southeast draining streams that join downslope with the southwest-northeast flowing Nura River. Most of the streams crossing the 2008 fault zone are disrupted, especially in areas where the topographic highs are most pronounced. In these sectors, the streams in the upper drainage areas have a trellis pattern and have adjusted to the ridges by flowing at right angles and parallel to them. Between kilometers 4 and 5 (Figure 4), in an area of approximately 0.5 square kilometers, deeper incised stream channels expose alternating Paleogene brecciated gypsum and limestone units that occur topographically both above and below the fault zone.

4.2 Surface Rupture Distribution and Geometry

While the surface rupture of the 2008 Nura earthquake extends over a distance of approximately 8 km, our analysis focused mainly on a 6-km-long section that is well documented on DSMs (Figure 5). The rupture trace follows an average bearing of 040°, subparallel to the terraced lateral moraine deposits along the Nura River. Interestingly, the rupture is generally aligned with local topographic highs that offset the northern valley flanks of the Nura River. Our systematic digital rupture-trace mapping revealed 801 surface breaks (traces) trending on average 040°, of which roughly 70% have an upslope, northwest-facing scarp (Figure 5). Field measurements of the opening direction indicated a local bearing range between 090° and 100°, confirming the left-lateral sense of motion suggested by Teshebaeva et al. (2014) (Figure S1 in Supporting Information S1). The longest continuous break was 195 m long, and the distribution of surface breaks spanned a width of about 150 m at some locations (Figure 4). At around kilometer 4, the fault trace diverges from the general strike by about 250 m and follows a double bend that coincides with the Paleogene gypsum outcrop (compare Figure 2). The observed surface breaks exhibit a wide variety of surficial deformation patterns along strike. Based on characteristic tectono-geomorphic and structural features—including the width of the damage zone, the orientation of surface breaks relative to the overall trend of the rupture zone, and their relationship with local topographic highs—the rupture trace can be divided into four distinct zones (A–D) from northeast to southwest (Figure 5).

The northeastern terminus of the rupture in Zone A lies beyond the boundaries of the well-located DSMs (Figure S3 in Supporting Information S1). The surface rupture here is confined to a narrow band of discontinuous surface breaks only a few meters long that cut across a planar topographic surface (Figure 5). With little to no vertical offset and a decimeter-wide horizontal opening, the breaks form fissures that are several centimeters to decimeters deep. Most of Zone A is characterized by a straight, northeast-oriented sector of extensional fractures that follow the crests of local topographic highs. A series of small, northwest-facing offsets was first observed in the DSMs located tens of meters upslope of the principal extension zone, where no unambiguously identifiable topographic highs exist (kilometers 0–2; Figure 5a). In the center of this sector, the fault zone widens to ∼100 m, forming a narrow lens-shaped area in map view. Here, the local topography is less pronounced compared to the rest of Zone A, and the surface breaks form ∼50-cm-deep and up to ∼30-m-wide local graben (kilometers 1–1.5; Figures 5 and 6). Toward the southwestern end of Zone A (kilometers 1.5–2; Figure 5), the surface breaks become more prominent and continuous, forming extensional breaks associated with collapsed scarp flanks along the northwest-facing slopes of the several-meter-high topographic highs close to their crestal sectors (Figure 6).

The transition between Zones A and B is characterized by two right-stepping, several-meter-high topographic highs that deviate southward from the overall 040 trend, downslope toward the Nura River (Figure 5b). Both structures are connected at their northwestern terminus by a trace of 2008 surface breaks that offsets a flat surface by <50 cm, similar to the offsets observed upslope in Zone A. The more prominent 2008 breaks here run along the west-facing scarp face of the topographic highs, including less continuous, shorter, and in some cases en échelon tensile cracks that broke boulders in a dextral manner (Figure 6).

In Zone B, the fault zone widens again to form another lens-shaped sector in map view. This feature is approximately 150 m wide and 1,000 m long (kilometers 3–4; Figure 5). The local relief in this sector is less pronounced compared to the rest of the rupture zone and is similar to that observed in the center of Zone A. However, in contrast to Zone A, here the surface-rupture splays within the lens form a step-like pattern that is oblique with respect to the main rupture trace. These breaks are characterized by a densely spaced series of near-vertical, sub-parallel, west-facing scarps up to 1 m high, some of which are flanked by collapsed scarps and hanging-wall blocks (Figure 6). In places, the scarps are part of fissures up to 120 cm deep. At the southwestern end of Zone B (kilometer 3.8; Figure 5), topsoil layers are ripped open in a fan shape over a bulging local topography (Figure 6). Similar to Zone A and the transition zone, the northwestern boundary of the fault zone in Zone B again comprises less prominent, northwest-facing fault scarps that offset a flat surface.

Between kilometers 4 and 5, Zone C hosts the most prominent topographic highs along the entire fault zone, with well-expressed scarps up to 10 m high. The surface rupture in this zone is confined to a narrow strip, only a few meters wide, where the front of the northwest-facing limb of the topographic high is fractured. Here, Paleogene gypsum layers dipping 60°–80° to the southeast and mollusk-bearing limestone breccias that appear to have guided the deformation are observed (Figure 6). The gypsum units resurface below the fault zone, exhibiting a reversed dip of 25°–30° to the northwest.

Beyond kilometer 5, at the southwestern terminus, the Nura surface rupture bifurcates (Zone D; Figure 5). The upper (northwestern) splay was first discovered during digital mapping and can be traced over a length of approximately 400 m, following the orientation of the rupture in Zone C. However, because the DSM terminates in this area, the continuation of the rupture beyond this point is uncertain. The second splay deviates ∼20° southward (counterclockwise) and extends along strike for another ∼1,200 m. Similar to the breaks from the northeastern terminus in Zone A, the surface breaks in Zone D become less pronounced and more discontinuous, with gaps of tens to hundreds of meters. Around kilometer 6, there is a prevalent 150-m-long southeast-facing vertical scarp that reaches offsets of up to 80 cm. This unique feature stands out from the prominent northwest-facing upslope offsets found throughout the fault zone.

4.3 Vertical Separation Measurements

In the following, we combine field measurements with our DSM data to present a comprehensive analysis of VS measurements with their associated uncertainties along the surface rupture. The field-based displacement-vector measurements were obtained at the most pronounced surface offsets (Figure S1 in Supporting Information S1). However, an examination of the surface rupture within the DSM revealed that the vertical offset values obtained during the field campaign did not accurately represent the overall magnitude of net VS, particularly in regions characterized by spatially distributed faulting. To develop a more comprehensive understanding of the VS distribution along the Nura 2008 rupture, we applied a systematic modeling approach using topographic profiles across the rupture zone as well as longer topographic profiles across topographic highs associated with the 2008 rupture zone extracted from the DSM. Our analysis of VS thus focused on the ∼6-km-long section of the Nura surface rupture that lies within the area of the DSM (Figure 7).

With this approach we were able to compile a database of 303 VS measurements along the 2008 rupture zone and 56 VS measurements of associated topographic highs obtained from the DSM (Figure 7). When comparing the VS measured remotely with the VS data obtained in the field, we find that the measurements in Zone A, the transition zone, and Zone D align well within the uncertainties. In Zone B, the field measurements underestimate the remotely measured net slip by ∼0.5–1.5 m, which is due to the fault-splay distribution not accounted for in the field. The VS net slip along northwest-facing surface breaks range from 0.2 to 2.1 m (∼0.1 to 3.3 m including uncertainties). The resulting average VS_ave along the 6-km-long section of the rupture is 0.6 ± 0.2 m. The topographic highs range from 0.4 to 9.8 m (Figure 7). The distribution of VS along the 2008 rupture displays three prominent peaks. Two of these peaks are located within Zone A, between kilometers 0 to 1 and 1 to 2, respectively, and the third, most pronounced peak coincides with Zones B and C, between kilometers 3 and 5 (Figure 7). Notably, the peaks increase gradually from northeast to southwest, with the increase appearing to be approximately proportional to the increasing width of each VS distribution envelope. On average, the overall geometry of the distribution envelope leans asymmetrically toward the southwest. Interestingly, the VS distribution along the topographic highs exhibits three distinct peaks that appear to align with the positions of the peaks observed in the 2008 rupture distribution. However, unlike the 2008 distributions, the cumulative VS distribution peaks of the tectonic landforms have approximately the same maxima.

5 Discussion

Our new structural, stratigraphical, and geomorphological observations of the surface manifestations of the 2008 Nura earthquake and analysis of the spatial characteristics of vertical displacements in the rupture zone provide information on the geometry and kinematics of the seismogenic structures activated during this event. Below, we review the newly acquired data and evaluate them with respect to deformation models derived from previous geophysical and seismological investigations, followed by a comparison to similar structural settings elsewhere. When put into context with the local topographic conditions and the regional structural framework, this information contributes to understanding the long-term behavior of the IrkF as an integral structure of the Pamir—Tien Shan collision zone.

5.1 Rupture Kinematics

Using InSAR and seismological data, Teshebaeva et al. (2014) and Sippl et al. (2014), respectively, proposed deformation estimates and kinematic models to relate the surface rupture along the IrkF to the Nura main shock along the PFT. The two tectonic models proposed offer slightly different explanations for the structural and geomorphic features observed in the study area. For example, Teshebaeva et al. (2014) suggested a shared décollement and coseismic activation of the PFT and the IrkF during the rupture. To explain the missing seismicity along the IrkF, these authors also indicated that aseismic slip may have occurred in the Paleogene gypsum-bearing strata along the fault plane. In contrast, Sippl et al. (2014) suggested that the IrkF is associated with a ramp anticline located within the Paleozoic basement rocks and that the surface rupture is a result of uplift at its frontal limb. Both models share the characteristic that they are based on a deep-seated tectonic structure that was activated by the main shock along the PFT. We base our interpretation on these studies.

From the two studies mentioned above, it is known that the surface-rupture zone is associated with uplift accompanied by seismogenic events dominated by thrust and reverse faulting mechanisms. The resulting surface rupture, however, is characterized by extensional features. For example, at the northeastern terminus of the rupture zone, Zone A features extensional en échelon fractures, shallow grabens, and half grabens oriented subparallel to the surface rupture. Zone B, in contrast, is cut by near-vertical offsets, and exhibits grabens and fissures that are subparallel to one another but oriented obliquely relative to the overall trace of the rupture zone. A less prominent but important feature of the 2008 Nura rupture in Zones A and B and the connecting transfer zone is a narrow bend with small, northwest-facing scarps to the northwest and ahead of the principal deformation zone (PDZ) (see Figures 4 and 5). A similar tectono-geomorphic setting was observed in the 1988 M_w 6.8 Spitak earthquake in Armenia, where pressure ridges formed in frontal sectors of the uplifted hanging wall (Cisternas et al., 1989; Philip et al., 1992). In one of these cases, the collapse of the thrust front led to the detachment of the soil layer in the footwall and the formation of a shallow fold, referred to as a low-angle pressure ridge (Philip et al., 1992, their Figure 13), a feature resembling what we observed at Nura. The most striking extensional features characterize Zone C, an extensive area where Paleogene gypsum-bearing layers and resistant banks of limestone are exposed topographically above, within, and below the PDZ. Here, these strata dip south and northward between 60°–85° and 25°–30° (Figure 8), respectively, and we infer that these units form an integral part of the asymmetric syncline with a steep northern limb that is exposed north of the Nura settlement (see Figure 2). The southwestern end of the rupture Zone D is cut by surface breaks that gradually become shorter and less continuous before eventually dying out. Throughout these zones, but more prominently in Zone A and especially in Zone C, the rupture aligns well with the crests of local topographic highs within the glacial till of the northern valley flank of the Nura River, a relationship that will be discussed in Section 5.4. Assuming that the bedding exposed in Zone C continues laterally beneath the glacial till in Zone D, we hypothesize that the existence of the steeply inclined strata causes a hydrological barrier that causes the ponding of water and its localized upward movement along this part of the fault zone. This could explain the ubiquitous paludal environments in this area (e.g., Jones & Deike, 1981) and the existence of collapse structures along the inferred fault zone similar to fault-line sinkholes observed within the Pamir highland (Strecker et al., 1995) and elsewhere (e.g., Riedl et al., 2020).

5.2 Fault Model for the 2008 Nura Surface Rupture

To develop a fault model for the Nura earthquake zone, we compare our observation of the structural characteristics of the fault zone with similar phenomena related to seismogenic landforms, and associated deformation patterns during ruptures elsewhere. The landforms and deformation patterns we observed along the 2008 Nura surface rupture resemble those observed in regions characterized by shortening associated with flexural-slip faulting. Flexural-slip faulting is a phenomenon in which the deformation of folded sequences leads to sliding along mechanically weaker bedding surfaces that separate more rigid, mechanically stronger layers (Li et al., 2015, 2017; Yeats, 1986; Yeats et al., 1981). In areas where these folds are unconformably overlain by sediments, the movement is transmitted to the surface, causing these sediments to rupture. For instance, during the M_w ∼ 6 1994 Sefidabeh earthquake sequence in Iran, slip was accommodated along subvertical bedding planes within a fold that was situated above a seismically activated deep seated thrust fault, resulting in surface faulting along a ridge crest similar to what is seen in Zone A (e.g., Berberian et al., 2000). The oblique arrangement of structures in Zone B corresponds well with the sinistral component estimated by Teshebaeva et al. (2014), and is analogous to other regions characterized by oblique shortening. For example, during the 1980 M_s 7.3 El Asnam earthquake (Algeria), bending of the surface generated grabens that deviate from the strike of the rupture zone at the crest of topographic highs (e.g., Philip & Meghraoui, 1983). However, at Nura the interbedding of breccias within both the folded units and the fault zone in Zone C is the strongest indicator that flexural-slip faulting was the driving force of the surface rupture within the overlying glacial till.

A similar tectonic scenario characterizes the sedimentary strata in the Tarim Basin immediately east of our study area. There, analogous relationships exist between folded Neogene sedimentary strata and the deformed Quaternary gravels and landforms above them. In the Tarim Basin, Quaternary deformation is associated with the growth of flexural-slip folds in the vicinity of thrust faults that are either linked with or result from movement along detachments and ramps located only a few kilometers beneath the surface (e.g., Bufe et al., 2017; Li et al., 2018, 2019; Scharer et al., 2004, 2006; J. A. Thompson et al., 2015; Thompson Jobe et al., 2017). Presuming a comparable tectonic configuration in the Nura area and drawing on the analogous ruptures mentioned earlier, we posit that the movement along the IrkF associated with the 2008 Nura earthquake is due to internal deformation of Cenozoic folds that unconformably rest upon folded Paleozoic rocks, which in turn are bounded by the PFT (Figure 9). Consequently, the 2008 surface rupture can be attributed to bedding-parallel flexural-slip faulting along the southeast-dipping limb of the syncline overlain by Quaternary glacial till (Figure 8). The variations in the observed faulting patterns at surface could be related to (a) variations in the thickness of the overlying till, ranging from a few to tens of meters in some areas to nearly no preserved cover sediments in other areas (Figure 8); (b) irregularities in the geometry of the underlying fold; (c) varying thicknesses of mechanically weaker bedding surfaces; and (d) the character (energy and frequency content) of the passing dynamic seismic wave field.

The next step is to contextualize our results within the seismological and geodynamic framework of the Nura earthquake and its aftershocks. The understanding of flexural-slip faulting has improved significantly in recent decades. Corresponding studies have shown that while flexural-slip faults typically stay within the bedding of sedimentary cover rocks and do not nucleate deeper than the amplitude of the flexural-slip fold, they might be driven to slip indirectly by the activity of deep-seated or nearby bounding faults (Figure 9) (e.g., Kelsey et al., 2008; Philip & Meghraoui, 1983; Rockwell et al., 2014; Yeats, 1986). However, the occurrence of flexural-slip faulting accompanying earthquakes has been only observed in few cases; for instance, the 1944 7 M_w San Juan earthquake in western Argentina (Rockwell et al., 2014), the 1980 M_w 7.3 El Asnam earthquake in Algeria (Philip & Meghraoui, 1983), or the 2013 M_w 6.6 Lake Grassmere earthquake in New Zealand (Kaneko et al., 2015). In the first two cases, slip along a first-order reverse fault was accompanied by extensive internal deformation of the surrounding rocks, which caused flexural-slip folding in adjacent folds and flexural-slip faulting on bedding planes, which produced fault scarps with surface displacement away from the causative fault (Figure 9a). In contrast, the study of the Lake Grassmere earthquake found that the active reverse fault bounding the fold did not slip during the earthquake, and Kaneko et al. (2015) propose that motion along the flexural-slip faults was likely caused by dynamic shaking during seismic wave passage. This suggests that flexural-slip faults may respond to dynamic stresses related to seismic waves, as proposed by Teshebaeva et al. (2014) for the Nura earthquake and also described for triggering of shallow slip elsewhere (Victor et al., 2018). In view of the regional geological context in our study area, we infer that the deformation may be a result of either (a) movement along the PFT, (b) transmission of dynamic stresses from seismic waves, (c) slip along a buried thrust triggered by the main shock, or a combination of these factors.

5.3 Understanding Flexural-Slip Faulting at the Nura Site: Scaling Relationships and Discrepancies

Detailed documentation of VS, surface-rupture length (SRL), and faulting characteristics is a valuable database for the Worldwide and Unified Database of Surface Ruptures (SUrface Ruptures due to Earthquakes [SURE]) and Fault Displacement Hazard Analysis (FDHA) (Baize et al., 2019). The FDHA relies on empirical relationships established from historical earthquake fault ruptures, constrained by a data set with limited representation across diverse tectonic contexts. This data set covers only a specific magnitude range (M > 6.5) and disregards crucial parameters such as surface geology and structural complexity (Baize et al., 2019). A specific concern in this regard is the inadequate coverage of widely distributed or secondary deformation features, such as flexural-slip ruptures associated with coseismic fault-related folding (Baize et al., 2019). Established in 2015, SURE addresses these limitations, aiming to enhance FDHA robustness and improve empirical scaling relationships between earthquake magnitude and fault-displacement parameters for probabilistic FDHA (PFDHA). Our study contributes new information on remotely triggered surface ruptures by providing three-component files describing the earthquake, rupture section, and slip observations that were standardized under SURE, and subsequently uploaded to the SURE 2.0 database (Nurminen et al., 2022).

Using the SURE 2.0 nomenclature (Nurminen et al., 2022), we classify our surface rupture as a triggered distributed (rank 3) and flexural-slip (F-S, rank 22) event. Our investigation, which was conducted a decade after the 2008 Nura rupture had occurred, focused on a 6-km-long segment of the total 8 km surface rupture owing to limited DSM availability. We determined an VS_ave of 0.6 + 0.2/−0.2 m, with a maximum VS_max of 2.1 + 1.2/−0.5 m. These values are generally in line with the 1.6–2 m estimated by Teshebaeva et al. (2014), who used an optimal fault model to calculate vertical offset within the earthquake rupture zone. Despite the passage of a decade since the rupture event and our investigations, we maintain confidence in our VS measurements of nearly vertical offsets due to their reliance on modeled projections of the offset far-field surface, which remains largely unaffected by erosion. Conversely, measuring the strike-slip offset depends on identifying clearly separated reference points along the edges of the surface rupture scarp. However, erosion has considerably blurred potential points over the last 10 years, making accurate measurements difficult. Nevertheless, we observed consistent indications of strike-slip movement along fractured, laterally offset boulders.

The use of scaling relationships with our data set with respect to fault displacement is complicated by several limitations. First, our measurements primarily represent VS, that is, the vertical difference between two planar markers across the surface rupture. This contrasts with established studies of the scaling relationships that consider displacement (i.e., dip slip), which represents the relative motion between initially adjacent points along the fault plane (e.g., Moss & Ross, 2011; Wells & Coppersmith, 1994). Second, VS measurements made in the glacial till cover of the folded strata capture the translation of slip at the surface and might not fully represent the actual slip along the flexural faults. Third, it remains difficult to establish a link between the faulting phenomena observed at the surface and the slip mechanisms at depth, and the potential dip of the buried fault. Finally, an important caveat lies in our currently insufficient understanding of the driving forces behind the observed flexural-slip faults, particularly in the context of distinguishing whether faulting along these structures occurred simultaneously with the Nura main shock or whether it was associated with low-magnitude aftershocks and/or afterslip.

To further examine the latter aspect, we performed regression analyses to establish correlations between SRL and moment magnitude using the equation M_w = 5.08 + 1.16*log(SRL) for all faulting mechanisms established by Wells and Coppersmith (1994). This regression yields a minimum magnitude of M_w 6.1 for a SRL of 8 km. According to this scaling relationship, a magnitude of M_w 6.6, such as that of the 2008 Nura earthquake, and the M_w 6.5 estimated by Teshebaeva et al. (2014) from their fault model would correspond to a SRL of approximately 20 km. This suggests that the observed surface rupture might be shorter than expected, indicating potential complexities in the faulting mechanisms or other factors influencing the observed rupture lengths. Indeed, studies of flexural-slip faulting in other areas emphasized the profound influence of geological conditions and fold geometry on the relation between earthquake magnitude and SRL that does not follow scaling relationships established from normal, thrust and strike-slip faulting mechanisms. For example, the 1981 M_L 2.5 (M_w ≈ 2.6) Lompoc earthquake yielded an SRL of almost 600 m despite its low magnitude (Yerkes et al., 1983). Conversely, the 2013 M_w 6.6 Lake Grassmere earthquake triggered three flexural-slip faults, each with a SRL of up to 1.5 km (Kaneko et al., 2015), while the 1944 M_w 7.0 San Juan earthquake resulted in a SRL of 7 km. Both latter cases exhibited rupture lengths significantly shorter than those predicted by scaling relationships for their respective magnitudes. Because it is uncertain whether the surface rupture was coeval with the main shock, it is also possible that it resulted from a lower magnitude aftershock. According to Sippl et al. (2014), the largest aftershocks in the sequence were in the magnitude range of 5.4. Given this magnitude, a SRL of 8 km would be unlikely. However, as proposed by Teshebaeva et al. (2014), the presence of evaporites could have reduced friction along the fault by acting as a lubricant and facilitating slip. This, in turn, could have led to locally elevated displacements and caused a greater expansion of SRL in the overlying glacial till, despite a lower magnitude obtained from the aftershocks.

5.4 Long-Term Geomorphic Evolution of the Irkeshtam Fault and Hazard Assessment

The 2008 Nura surface rupture appears to co-vary consistently with local topographic highs where it ruptures along the ridge crest or breaks up the northwest-facing limb of the fault-parallel fold in the Paleogene sedimentary rocks. Notably, these topographic highs are mostly associated with well-developed ponding, abandoned, and in some cases deflected stream channels. The geomorphic characteristics strongly suggest that the topographic highs represent cumulative tectonic landforms that formed during earlier, probably recurrent ruptures and/or deformation in this zone. To unravel the history of these tectonic landforms, it is crucial to consider the regional tectonic setting of the Nura area and determine both the onset of faulting and the rate of deformation.

The Nura area is characterized by complete closure of the topographic depression between the Pamir and Tien Shan, resulting in deformation, uplift, and exposure of Paleozoic to Cenozoic units. This history of uplift and deformation is further evident in the major drainage divide between the catchment areas of the present-day Kyzilsu rivers (westward- and eastward-flowing, respectively) that formed during the late Cenozoic (Pavlis et al., 1997; Strecker et al., 2003). The Nura River cuts through valley moraines and is an integral part of the eastern Kyzilsu catchment. The remarkable deflection of both the past glaciers and resulting moraines, and the Nura River against the IrkF indicates that the course of the river was determined by both, activity along the IrkF and uplift during the progressive closure of the basin. This suggests that the activity of the IrkF and the development of tectonically forced drainage patterns predate the deposition of glacial deposits along the river. Consequently, an assessment of faulting activity can be made within the age range of the deposits where deformation is evident in the form of tectonic landforms along the IrkF, that is, in the Qm2 glacial deposits. As many of the complex details of the basin-closure history were likely obscured by the combined erosional effects of preceding glaciations and protracted tectonic activity, longer-term assessments are difficult to make in the absence of further geochronological data.

Despite these problems, the geochronological control from the Nura site does provide two important age estimates to constrain the time period when the tectonic landforms associated with the 2008 surface rupture began to form. First, an IRSL date obtained from the reworked loess covering the Qm2 glacial deposits within the 2008 rupture zone provides a minimum age estimate of 11.3 ± 0.5 ka. Second, on the northeastern side of the Nura River, an IRSL age of 12.0 ± 0.6 ka was determined from lacustrine sediments north of the Nura settlement. The paleo-lake environment was associated with a mass-movement deposit (L1) that was dislodged from the flank of an open syncline by a landslide. Based on regional structural relationships, we infer that the flank of the syncline is the northeastern extension of the fold covered by the Qm2 glacial deposits that were affected by the 2008 fault zone. Assuming that our interpretation of the age relationships and the glacial depositional record is accurate, the IRSL ages indicate a time span of approximately 12 kyr for the growth of the IrkF-related relief. Combined with our estimated cumulative VS_ave (4.6 m), this results in a maximum Holocene offset rate of 0.4 mm/yr. In addition, the correlation between the distribution of the displacements in 2008 and the cumulative displacement patterns (Figure 7) strongly suggests that earthquakes in the regional fault system that triggered similar surface ruptures must have occurred repeatedly in the past. If this is true, the cumulative VS_ave divided by the 2008 VS_ave (0.64 m) indicates that about seven earthquakes similar to the 2008 Nura event occurring at potential (minimum) intervals of approximately 1.7 kyr would have been necessary to generate the present-day fault-zone morphology since the formation of those markers. Although these assessments should be viewed with great caution due to the limited availability of robust geochronological data, it is interesting to note that these findings are compatible with the estimated recurrence interval of 1.9 kyr along the central segment of the PFT, which is characterized by complete tectonic segment activation and potential segment interaction during great M > 7 earthquakes (Patyniak et al., 2021). However, to validate our estimates, further paleoseismic investigations along the IrkF are necessary.

6 Conclusions

In this study, we conducted detailed structural, stratigraphical, and geomorphological analyses of the 2008 M_w 6.6 Nura earthquake rupture zone in the Northern Pamir continental collision zone in Central Asia. Our primary aim was to understand the relevant seismogenic structures and their geometry, kinematics, and influence on the formation of tectono-geomorphic fault-zone characteristics. Our investigation built on existing fault-zone models proposed by Teshebaeva et al. (2014) and Sippl et al. (2014), and revealed flexural-slip faulting to be the driving mechanism for the observed surface-rupture patterns. Our main findings are:

The 2008 Nura earthquake surface rupture resulted from bed-parallel flexural-slip faulting along the southeast-dipping limb of a syncline in Paleogene sedimentary strata that are overlain by unconsolidated Quaternary glacial till. We suggest that the earthquake was either triggered coseismically by rupture along a concealed fault or by slip following the main event associated with dynamic stresses induced by seismic wave passage.
Both the magnitude and nature of surface deformation were intricately shaped by variations in till thickness, potential irregularities in fold geometry, and geological conditions, making it difficult to accurately determine scaling relationships.
The study provides new information for the Worldwide and Unified Database of Surface Ruptures (SURE), classifying the rupture as a triggered distributed, flexural-slip event.

Our findings elucidate the long-term behavior of the IrkF, as indicated by the correlation between 2008 surface rupture and local tectonic landforms. Geochronological data suggest a potential recurrence interval of approximately 1.7 kyr for earthquakes similar to the 2008 Nura event during the Holocene. However, further paleoseismic investigations are required to validate these estimates. The study provides valuable insights into the Nura fault model, fault displacement, and hazard assessment, and highlights the complexities of seismogenic flexural-slip faulting and the need for continued research to enhance our understanding of remotely triggered faults.

Global Research Collaboration Statement

We extend our sincere gratitude to the Institute of Seismology at the National Academy of Sciences of Kyrgyzstan (Bishkek, Kyrgyzstan) for their invaluable partnership and support during our field campaign. Their provision of resources, local expertise, guidance, and on-the-ground assistance was instrumental to the success of our work.

Acknowledgments

This project is part of the CaTeNA-project within the Client II program of and funded by the German Federal Ministry of Education and Research (BMBF; Sub-project Grant 03G0878E to M. Strecker). We would like to thank J. Mosar and E. Sobel for sharing their observations and the insightful discussions, which enhanced our knowledge and understanding of the study area. We would also like to thank F. Wang for sharing geological observations from the Chinese side of our study area. We are grateful to S. Bello, one anonymous reviewer and associate editor J. Bruce H. Shyu for their constructive comments that improved this manuscript. Open Access funding enabled and organized by Projekt DEAL.

Open Research

Data Availability Statement

All TanDEM-X data were kindly provided by the German Aerospace Center (DLR) for scientific use. Basemap data for the overview maps was sourced from OpenTopography. The data produced in this study is stored in open-access online archives. UAV videos, shapefile, field measurements and associated photographs from the 2008 surface rupture are available at Zenodo http://doi.org/10.5281/zenodo.7961956 (Patyniak, 2024). Raster-tiff data of the digital surface model is available at OpenTopography https://doi.org/10.5069/G9ZW1J4C (Patyniak et al., 2024).

Supporting Information

References

Abdrakhmatov, K. E., Omuraliev, M., & Ormukov, C. (2008). Nura earthquake October 5, 2008 with КР = 15.4, M_w = 6.7, I0 = 8 (Kyrgyzstan). Izvestiya (No. 4).
Google Scholar
Arrowsmith, J R., & Strecker, M. R. (1999). Seismotectonic range-front segmentation and mountain-belt growth in the Pamir-Alai region, Kyrgyzstan (India-Eurasia collision zone). GSA Bulletin, 111(11), 1665–1683. https://doi.org/10.1130/0016-7606(1999)111<1665:SRFSAM>2.3.CO;2
10.1130/0016-7606(1999)111<1665:SRFSAM>2.3.CO;2
ADSWeb of Science®Google Scholar
Baize, S., Nurminen, F., Sarmiento, A., Dawson, T., Takao, M., Scotti, O., et al. (2019). A worldwide and unified database of surface ruptures (SURE) for fault displacement hazard analyses. Seismological Research Letters, 91(1), 499–520. https://doi.org/10.1785/0220190144
10.1785/0220190144
Web of Science®Google Scholar
Bello, S., Scott, C. P., Ferrarini, F., Brozzetti, F., Scott, T., Cirillo, D., et al. (2021). High-resolution surface faulting from the 1983 Idaho Lost River Fault M_w 6.9 earthquake and previous events. Scientific Data, 8(1), 68. https://doi.org/10.1038/s41597-021-00838-6
10.1038/s41597-021-00838-6
PubMedGoogle Scholar
Bemis, S. P., Micklethwaite, S., Turner, D., James, M. R., Akciz, S. O., Thiele, S. T., & Bangash, H. A. (2014). Ground-based and UAV-based photogrammetry: A multi-scale, high-resolution mapping tool for structural geology and paleoseismology. Journal of Structural Geology, 69, 163–178. https://doi.org/10.1016/j.jsg.2014.10.007
10.1016/j.jsg.2014.10.007
ADSWeb of Science®Google Scholar
Berberian, M., Jackson, J. A., Qorashi, M., Talebian, M., Khatib, M., & Priestley, K. (2000). The 1994 Sefidabeh earthquakes in eastern Iran: Blind thrusting and bedding-plane slip on a growing anticline, and active tectonics of the Sistan suture zone. Geophysical Journal International, 142(2), 283–299. https://doi.org/10.1046/j.1365-246x.2000.00158.x
10.1046/j.1365-246x.2000.00158.x
ADSWeb of Science®Google Scholar
Bilham, R., Gaur, V. K., & Molnar, P. (2001). Himalayan seismic hazard. Science, 293(5534), 1442–1444. https://doi.org/10.1126/science.1062584
10.1126/science.1062584
CASPubMedWeb of Science®Google Scholar
Bufe, A., Bekaert, D. P. S., Hussain, E., Bookhagen, B., Burbank, D. W., Thompson Jobe, J. A., et al. (2017). Temporal changes in rock uplift rates of folds in the foreland of the Tian Shan and the Pamir from geodetic and geologic data. Geophysical Research Letters, 44(21), 10977–10987. https://doi.org/10.1002/2017GL073627
10.1002/2017GL073627
ADSWeb of Science®Google Scholar
Burtman, V. S., & Molnar, P. (1993). Geological and geophysical evidence for deep subduction of continental crust beneath the Pamir. Geological Society of America.
10.1130/SPE281-p1
Google Scholar
Chapman, J. B., Carrapa, B., Ballato, P., DeCelles, P. G., Worthington, J., Oimahmadov, I., et al. (2017). Intracontinental subduction beneath the Pamir Mountains: Constraints from thermokinematic modeling of shortening in the Tajik fold-and-thrust belt. GSA Bulletin, 129(11–12), 1450–1471. https://doi.org/10.1130/B31730.1
10.1130/B31730.1
Google Scholar
Cisternas, A., Philip, H., Bousquet, J. C., Cara, M., Deschamps, A., Dorbath, L., et al. (1989). The Spitak (Armenia) earthquake of 7 December 1988: Field observations, seismology and tectonics. Nature, 339(6227), 675–679. https://doi.org/10.1038/339675a0
10.1038/339675a0
ADSWeb of Science®Google Scholar
Coutand, I., Strecker, M. R., Arrowsmith, J R., Hilley, G. E., Thiede, R., Korjenkov, A., & Omuraliev, M. (2002). Late Cenozoic tectonic development of the intramontane Alai Valley, (Pamir-Tien Shan region, central Asia): An example of intracontinental deformation due to the Indo-Eurasia collision. Tectonics, 21(6), 3-1–3-19. https://doi.org/10.1029/2002TC001358
10.1029/2002TC001358
ADSWeb of Science®Google Scholar
Degering, D., & Degering, A. (2020). Change is the only constant—Time-dependent dose rates in luminescence dating. Quaternary Geochronology, 58, 101074. https://doi.org/10.1016/j.quageo.2020.101074
10.1016/j.quageo.2020.101074
Web of Science®Google Scholar
DuRoss, C. B., Bunds, M. P., Gold, R. D., Briggs, R. W., Reitman, N. G., Personius, S. F., & Toké, N. A. (2019). Variable normal-fault rupture behavior, northern Lost River fault zone, Idaho, USA. Geosphere, 15(6), 1869–1892. https://doi.org/10.1130/GES02096.1
10.1130/GES02096.1
ADSWeb of Science®Google Scholar
Dziewonski, A. M., Chou, T.-A., & Woodhouse, J. H. (1981). Determination of earthquake source parameters from waveform data for studies of global and regional seismicity. Journal of Geophysical Research, 86(B4), 2825–2852. https://doi.org/10.1029/JB086iB04p02825
10.1029/JB086iB04p02825
ADSWeb of Science®Google Scholar
Ekström, G., Nettles, M., & Dziewoński, A. M. (2012). The global CMT project 2004–2010: Centroid-moment tensors for 13,017 earthquakes. Physics of the Earth and Planetary Interiors, 200–201, 1–9. https://doi.org/10.1016/j.pepi.2012.04.002
10.1016/j.pepi.2012.04.002
ADSWeb of Science®Google Scholar
Elliott, J. R., Jolivet, R., González, P. J., Avouac, J.-P., Hollingsworth, J., Searle, M. P., & Stevens, V. L. (2016). Himalayan megathrust geometry and relation to topography revealed by the Gorkha earthquake. Nature Geoscience, 9(2), 174–180. https://doi.org/10.1038/ngeo2623
10.1038/ngeo2623
CASADSWeb of Science®Google Scholar
Fan, G., Ni, J. F., & Wallace, T. C. (1994). Active tectonics of the Pamirs and Karakorum. Journal of Geophysical Research, 99(B4), 7131–7160. https://doi.org/10.1029/93JB02970
10.1029/93JB02970
ADSWeb of Science®Google Scholar
Galbraith, R. F., & Roberts, R. G. (2012). Statistical aspects of equivalent dose and error calculation and display in OSL dating: An overview and some recommendations. Quaternary Geochronology, 11, 1–27. https://doi.org/10.1016/j.quageo.2012.04.020
10.1016/j.quageo.2012.04.020
Web of Science®Google Scholar
Ge, J., Shi, X., Chen, H., Lin, X., Ge, W., Wei, X., et al. (2022). Two kinematic transformations of the Pamir salient since the Mid-Cenozoic: Constraints from multi-timescale deformation analysis. Frontiers in Earth Science, 10, 967529. https://doi.org/10.3389/feart.2022.967529
10.3389/feart.2022.967529
ADSGoogle Scholar
Gray, H. J., Mahan, S. A., Rittenour, T., Nelson, M., & Lund, W. R. (2015). Guide to luminescence dating techniques and their applications for paleoseismic research. In Proceedings volume, Basin and range province seismic hazards Summit III (p. 15-5). Utah Geological Survey Miscellaneous Publication.
Google Scholar
Hamburger, M. W., Sarewitz, D. R., Pavlis, T. L., & Popandopulo, G. A. (1992). Structural and seismic evidence for intracontinental subduction in the Peter the first range, central Asia. GSA Bulletin, 104(4), 397–408. https://doi.org/10.1130/0016-7606(1992)104<0397:SASEFI>2.3.CO;2
10.1130/0016-7606(1992)104<0397:SASEFI>2.3.CO;2
Google Scholar
Hauksson, E., Jones, L. M., & Hutton, K. (1995). The 1994 Northridge earthquake sequence in California: Seismological and tectonic aspects. Journal of Geophysical Research, 100(B7), 12335–12355. https://doi.org/10.1029/95JB00865
10.1029/95JB00865
ADSWeb of Science®Google Scholar
Jackson, J. A., Molnar, P., Patton, H., & Fitch, T. (1979). Seismotectonic aspects of the Markansu valley, Tadjikstan, earthquake of August 11, 1974. Journal of Geophysical Research, 84(B11), 6157–6167. https://doi.org/10.1029/JB084iB11p06157
10.1029/JB084iB11p06157
ADSWeb of Science®Google Scholar
Johnson, K., Nissen, E., & Lajoie, L. (2018). Surface rupture morphology and vertical slip distribution of the 1959 M_w 7.2 Hebgen Lake (Montana) earthquake from airborne lidar topography. Journal of Geophysical Research: Solid Earth, 123(9), 8229–8248. https://doi.org/10.1029/2017JB015039
10.1029/2017JB015039
ADSWeb of Science®Google Scholar
Johnson, K., Nissen, E., Saripalli, S., Arrowsmith, J R., McGarey, P., Scharer, K. M., et al. (2014). Rapid mapping of ultrafine fault zone topography with structure from motion. Geosphere, 10(5), 969–986. https://doi.org/10.1130/GES01017.1
10.1130/GES01017.1
ADSWeb of Science®Google Scholar
Jones, W. K., & Deike, G. H. (1981). A hydrogeologic study of the watershed of the national Fisheries center at Leetown, West Virginia. Environmental Data, Frankford WV.
Google Scholar
Kalmetieva, Z. A., Mikolaichuk, A. V., Moldobekov, B. D., Meleshko, A. V., Jantaev, M. M., Zubovich, A. V., & Havenith, H. B. (2009). Atlas of earthquakes in Kyrgyzstan. In CAIAG, Bishkek (Vol. 76).
Google Scholar
Kaneko, Y., Hamling, I. J., van Dissen, R. J., Motagh, M., & Samsonov, S. V. (2015). Insar imaging of displacement on flexural-slip faults triggered by the 2013 M_w 6.6 Lake Grassmere earthquake, central New Zealand. Geophysical Research Letters, 42(3), 781–788. https://doi.org/10.1002/2014GL062767
10.1002/2014GL062767
ADSWeb of Science®Google Scholar
Kelsey, H. M., Sherrod, B. L., Nelson, A. R., & Brocher, T. M. (2008). Earthquakes generated from bedding plane-parallel reverse faults above an active wedge thrust, Seattle fault zone. Geological Society of America Bulletin, 120(11–12), 1581–1597. https://doi.org/10.1130/B26282.1
10.1130/B26282.1
ADSWeb of Science®Google Scholar
Kozlov, V. V., & Artemov, A. V. (1964). Geological map of Kyrgyzstan, scale 1:200,000, Alai valley sheet J-43-II. All union geolog. TREST.
Google Scholar
Li, T., Chen, J., Thompson, J. A., Burbank, D. W., & Yang, X. (2015). Active flexural-slip faulting: A study from the Pamir-Tian Shan convergent zone, NW China. Journal of Geophysical Research: Solid Earth, 120(6), 4359–4378. https://doi.org/10.1002/2014JB011632
10.1002/2014JB011632
ADSWeb of Science®Google Scholar
Li, T., Chen, J., Thompson Jobe, J. A., & Burbank, D. W. (2017). Active flexural-slip faulting: Controls exerted by stratigraphy, geometry, and fold kinematics. Journal of Geophysical Research: Solid Earth, 122(10), 8538–8565. https://doi.org/10.1002/2017JB013966
10.1002/2017JB013966
ADSWeb of Science®Google Scholar
Li, T., Chen, J., Thompson Jobe, J. A., Burbank, D. W., Cheng, X., Xu, J., et al. (2018). Active bending-moment faulting: Geomorphic Expression, Controlling conditions, accommodation of fold deformation. Tectonics, 37(8), 2278–2306. https://doi.org/10.1029/2018TC004982
10.1029/2018TC004982
ADSWeb of Science®Google Scholar
Li, T., Chen, Z., Chen, J., Thompson Jobe, J. A., Burbank, D. W., Li, Z., et al. (2019). Along-strike and downdip segmentation of the Pamir Frontal Thrust and its association with the 1985 M_w 6.9 Wuqia earthquake. Journal of Geophysical Research: Solid Earth, 124(9), 9890–9919. https://doi.org/10.1029/2019JB017319
10.1029/2019JB017319
ADSWeb of Science®Google Scholar
Lukas, S., Graf, A., Coray, S., & Schlüchter, C. (2012). Genesis, stability and preservation potential of large lateral moraines of Alpine valley glaciers—Towards a unifying theory based on Findelengletscher, Switzerland. Quaternary Science Reviews, 38, 27–48. https://doi.org/10.1016/j.quascirev.2012.01.022
10.1016/j.quascirev.2012.01.022
ADSWeb of Science®Google Scholar
Molnar, P., & Ghose, S. (2000). Seismic moments of major earthquakes and the rate of shortening across the Tien Shan. Geophysical Research Letters, 27(16), 2377–2380. https://doi.org/10.1029/2000GL011637
10.1029/2000GL011637
ADSWeb of Science®Google Scholar
Moss, R. E. S., & Ross, Z. E. (2011). Probabilistic fault displacement hazard analysis for reverse faults. Bulletin of the Seismological Society of America, 101(4), 1542–1553. https://doi.org/10.1785/0120100248
10.1785/0120100248
Web of Science®Google Scholar
Nikonov, A. A., Vakov, A. V., & Veselov, I. A. (1983). Seismotectonics and earthquakes in the convergent zone between the Pamir and the Tien Shan. Nauka.
Google Scholar
Nöth, L. (1932). Geologische Untersuchungen im nordwestlichen Pamirgebiet und mittleren Transalai. D. Reimer, E. Vohsen.
Google Scholar
Nurminen, F., Baize, S., Boncio, P., Blumetti, A. M., Cinti, F. R., Civico, R., & Guerrieri, L. (2022). Sure 2.0—New release of the worldwide database of surface ruptures for fault displacement hazard analyses. Scientific Data, 9(1), 729. https://doi.org/10.1038/s41597-022-01835-z
10.1038/s41597-022-01835-z
PubMedGoogle Scholar
Patyniak, M. (2024). 2008 Nura earthquake surface rupture slip vector documentation [Dataset]. Zenodo. https://doi.org/10.5281/zenodo.10911882
10.5281/zenodo.10911882
Google Scholar
Patyniak, M., Landgraf, A., Dzhumabaeva, A., Baikulov, S., Williams, A. M., Weiss, J. R., et al. (2021). The Pamir Frontal Thrust fault: Holocene full-segment ruptures and implications for complex segment interactions in a continental collision zone. Journal of Geophysical Research: Solid Earth, 126(12), e2021JB022405. https://doi.org/10.1029/2021JB022405
10.1029/2021JB022405
ADSWeb of Science®Google Scholar
Patyniak, M., Williams, A., & Arrowsmith, R. (2024). Survey of the 2008 Nura earthquake surface rupture, Kyrgyzstan, 2018 [Dataset]. Distributed by OpenTopography. https://doi.org/10.5069/G9ZW1J4C
10.5069/G9ZW1J4C
Google Scholar
Pavlis, T. L., Hamburger, M. W., & Pavlis, G. L. (1997). Erosional processes as a control on the structural evolution of an actively deforming fold and thrust belt: An example from the Pamir-Tien Shan region, Central Asia. Tectonics, 16(5), 810–822. https://doi.org/10.1029/97TC01414
10.1029/97TC01414
ADSWeb of Science®Google Scholar
Philip, H., & Meghraoui, M. (1983). Structural analysis and interpretation of the surface deformations of the El Asnam earthquake of October 10, 1980. Tectonics, 2(1), 17–49. https://doi.org/10.1029/TC002i001p00017
10.1029/TC002i001p00017
ADSWeb of Science®Google Scholar
Philip, H., Rogozhin, E., Cisternas, A., Bousquet, J. C., Borisov, B., & Karakhanian, A. (1992). The Armenian earthquake of 1988 December 7: Faulting and folding, neotectonics and palaeoseismicity. Geophysical Journal International, 110(1), 141–158. https://doi.org/10.1111/j.1365-246X.1992.tb00718.x
10.1111/j.1365-246X.1992.tb00718.x
ADSWeb of Science®Google Scholar
Pickering, K. T., Koren, T. N., Lytochkin, V. N., & Siveter, D. J. (2008). Silurian–Devonian active-margin deep-marine systems and palaeogeography, Alai range, southern Tien Shan, central Asia. Journal of the Geological Society, 165(1), 189–210. https://doi.org/10.1144/0016-76492006-082
10.1144/0016-76492006-082
ADSGoogle Scholar
Preusser, F., Degering, D., Fülling, A., & Miocic, J. (2023). Complex dose rate calculations in luminescence dating of lacustrine and Palustrine sediments from Niederweningen, northern Switzerland. Geochronometria, 50(1), 28–49. https://doi.org/10.2478/geochr-2023-0003
10.2478/geochr-2023-0003
CASGoogle Scholar
Preusser, F., Ramseyer, K., & Schlüchter, C. (2006). Characterisation of low OSL intensity quartz from the New Zealand Alps. Radiation Measurements, 41(7), 871–877. https://doi.org/10.1016/j.radmeas.2006.04.019
10.1016/j.radmeas.2006.04.019
CASADSWeb of Science®Google Scholar
Qiao, X., Wang, Q., Yang, S., Li, J., Zou, R., & Ding, K. (2015). The 2008 Nura M_w 6.7 earthquake: A shallow rupture on the main Pamir thrust revealed by GPS and InSAR. Geodesy and Geodynamics, 6(2), 91–100. https://doi.org/10.1016/j.geog.2015.01.005
10.1016/j.geog.2015.01.005
Google Scholar
Qiao, X., Yu, P., Nie, Z., Li, J., Wang, X., Kuzikov, S. I., et al. (2017). The crustal deformation revealed by GPS and InSAR in the northwest corner of the Tarim Basin, northwestern China. Pure and Applied Geophysics, 174(3), 1405–1423. https://doi.org/10.1007/s00024-017-1473-6
10.1007/s00024-017-1473-6
ADSGoogle Scholar
Reimann, T., & Tsukamoto, S. (2012). Dating the recent past (<500 years) by post-IR IRSL feldspar—Examples from the north Sea and Baltic Sea coast. Quaternary Geochronology, 10, 180–187. https://doi.org/10.1016/j.quageo.2012.04.011
10.1016/j.quageo.2012.04.011
Web of Science®Google Scholar
Reznichenko, N. V., Andrews, G. R., Geater, R. E., & Strom, A. (2017). Multiple origins of large hummock deposits in Alai Valley, Northern Pamir: Implications for palaeoclimate reconstructions. Geomorphology, 285, 347–362. https://doi.org/10.1016/j.geomorph.2017.02.019
10.1016/j.geomorph.2017.02.019
ADSWeb of Science®Google Scholar
Riedl, S., Melnick, D., Mibei, G. K., Njue, L., & Strecker, M. R. (2020). Continental rifting at magmatic centres: Structural implications from the late Quaternary Menengai Caldera, central Kenya Rift. Journal of the Geological Society, 177(1), 153–169. https://doi.org/10.1144/jgs2019-021
10.1144/jgs2019-021
ADSWeb of Science®Google Scholar
Robinson, T. R., Davies, T. R. H., Reznichenko, N. V., & Pascale, G. P. D. (2015). The extremely long-runout Komansu rock avalanche in the Trans Alai range, Pamir Mountains, southern Kyrgyzstan. Landslides, 12(3), 523–535. https://doi.org/10.1007/s10346-014-0492-y
10.1007/s10346-014-0492-y
Web of Science®Google Scholar
Rockwell, T. K., Ragona, D. E., Meigs, A. J., Owen, L. A., Costa, C. H., & Ahumada, E. A. (2014). Inferring a thrust-related earthquake history from secondary faulting: A long rupture record of La Laja fault, San Juan, Argentina. Bulletin of the Seismological Society of America, 104(1), 269–284. https://doi.org/10.1785/0120110080
10.1785/0120110080
Web of Science®Google Scholar
Scharer, K. M., Burbank, D. W., Chen, J., & Weldon, R. J. (2006). Kinematic models of fluvial terraces over active detachment folds: Constraints on the growth mechanism of the Kashi-Atushi fold system, Chinese Tian Shan. GSA Bulletin, 118(7–8), 1006–1021. https://doi.org/10.1130/B25835.1
10.1130/B25835.1
Google Scholar
Scharer, K. M., Burbank, D. W., Chen, J., Weldon, R. J., Rubin, C. M., Zhao, R., & Shen, J. (2004). Detachment folding in the southwestern Tian Shan–Tarim foreland, China: Shortening estimates and rates. Journal of Structural Geology, 26(11), 2119–2137. https://doi.org/10.1016/j.jsg.2004.02.016
10.1016/j.jsg.2004.02.016
ADSWeb of Science®Google Scholar
Schurr, B., Ratschbacher, L., Sippl, C., Gloaguen, R., Yuan, X., & Mechie, J. (2014). Seismotectonics of the Pamir. Tectonics, 33(8), 1501–1518. https://doi.org/10.1002/2014TC003576
10.1002/2014TC003576
ADSWeb of Science®Google Scholar
Sippl, C., Ratschbacher, L., Schurr, B., Krumbiegel, C., Rui, H., Pingren, L., & Abdybachaev, U. (2014). The 2008 Nura earthquake sequence at the Pamir-Tian Shan collision zone, southern Kyrgyzstan. Tectonics, 33(12), 2382–2399. https://doi.org/10.1002/2014TC003705
10.1002/2014TC003705
ADSWeb of Science®Google Scholar
Sobel, E. R., Chen, J., Schoenbohm, L. M., Thiede, R., Stockli, D. F., Sudo, M., & Strecker, M. R. (2013). Oceanic-style subduction controls late Cenozoic deformation of the Northern Pamir orogen. Earth and Planetary Science Letters, 363, 204–218. https://doi.org/10.1016/j.epsl.2012.12.009
10.1016/j.epsl.2012.12.009
CASADSWeb of Science®Google Scholar
Strecker, M. R., Frisch, W., Hamburger, M. W., Ratschbacher, L., Semiletkin, S., Zamoruyev, A., & Sturchio, N. (1995). Quaternary deformation in the eastern Pamirs, Tadzhikistan and Kyrgyzstan. Tectonics, 14(5), 1061–1079. https://doi.org/10.1029/95TC00927
10.1029/95TC00927
ADSWeb of Science®Google Scholar
Strecker, M. R., Hilley, G. E., Arrowsmith, J R., & Coutand, I. (2003). Differential structural and geomorphic mountain-front evolution in an active continental collision zone: The northwest Pamir, southern Kyrgyzstan. GSA Bulletin, 115(2), 166–181. https://doi.org/10.1130/0016-7606(2003)115<0166:DSAGMF>2.0.CO;2
10.1130/0016-7606(2003)115<0166:DSAGMF>2.0.CO;2
Google Scholar
Teshebaeva, K., Sudhaus, H., Echtler, H., Schurr, B., & Roessner, S. (2014). Strain partitioning at the eastern Pamir-Alai revealed through SAR data analysis of the 2008 Nura earthquake. Geophysical Journal International, 198(2), 760–774. https://doi.org/10.1093/gji/ggu158
10.1093/gji/ggu158
ADSWeb of Science®Google Scholar
Thompson, J. A., Burbank, D. W., Li, T., Chen, J., & Bookhagen, B. (2015). Late Miocene northward propagation of the northeast Pamir thrust system, northwest China. Tectonics, 34(3), 510–534. https://doi.org/10.1002/2014TC003690
10.1002/2014TC003690
ADSWeb of Science®Google Scholar
Thompson, S. C., Weldon, R. J., Rubin, C. M., Abdrakhmatov, K. E., Molnar, P., & Berger, G. W. (2002). Late Quaternary slip rates across the central Tien Shan, Kyrgyzstan, central Asia. Journal of Geophysical Research, 107(B9), ETG7-1–ETG7-32. https://doi.org/10.1029/2001JB000596
10.1029/2001JB000596
Web of Science®Google Scholar
Thompson Jobe, J. A., Li, T., Chen, J., Burbank, D. W., & Bufe, A. (2017). Quaternary tectonic evolution of the Pamir-Tian Shan convergence zone, Northwest China. Tectonics, 36(12), 2748–2776. https://doi.org/10.1002/2017TC004541
10.1002/2017TC004541
ADSWeb of Science®Google Scholar
Victor, P., Oncken, O., Sobiesiak, M., Kemter, M., Gonzalez, G., & Ziegenhagen, T. (2018). Dynamic triggering of shallow slip on forearc faults constrained by monitoring surface displacement with the IPOC Creepmeter Array. Earth and Planetary Science Letters, 502, 57–73. https://doi.org/10.1016/j.epsl.2018.08.046
10.1016/j.epsl.2018.08.046
CASADSWeb of Science®Google Scholar
Wells, D. L., & Coppersmith, K. J. (1994). New empirical relationships among magnitude, rupture length, rupture width, rupture area, and surface displacement. Bulletin of the Seismological Society of America, 84(4), 974–1002. https://doi.org/10.1785/BSSA0840040974
10.1785/BSSA0840040974
Web of Science®Google Scholar
Xiong, W., Qiao, X., Liu, G., Chen, W., & Nie, Z. (2019). Coulomb stress evolution along the Kongur Extensional System since 1895 and present seismic hazard. Geodesy and Geodynamics, 10(1), 1–9. https://doi.org/10.1016/j.geog.2018.11.007
10.1016/j.geog.2018.11.007
Web of Science®Google Scholar
Yeats, R. S. (1986). Active faults related to folding. Active Tectonics, 63–79.
Google Scholar
Yeats, R. S., Clark, M. N., Keller, E. A., & Rockwell, T. K. (1981). Active fault hazard in southern California: Ground rupture versus seismic shaking. Geological Society of America Bulletin, 92(4), 189. https://doi.org/10.1130/0016-7606(1981)92<189:AFHISC>2.0.CO;2
10.1130/0016-7606(1981)92<189:AFHISC>2.0.CO;2
ADSWeb of Science®Google Scholar
Yerkes, R. F., Ellsworth, W. L., & Tinsley, J. C. (1983). Triggered reverse fault and earthquake due to crustal unloading, northwest Transverse Ranges. California Geology, 11(5), 287. https://doi.org/10.1130/0091-7613(1983)11<287:TRFAED>2.0.CO;2
10.1130/0091?7613(1983)11<287:TRFAED>2.0.CO;2
Google Scholar
Zubovich, A. V., Wang, X., Scherba, Y. G., Schelochkov, G. G., Reilinger, R., Reigber, C., et al. (2010). GPS velocity field for the Tien Shan and surrounding regions. Tectonics, 29(6), TC6014. https://doi.org/10.1029/2010TC002772
10.1029/2010TC002772
ADSWeb of Science®Google Scholar

References From the Supporting Information

Volume43, Issue9

September 2024

e2024TC008360

Surface Rupture of the 2008 M_w 6.6 Nura Earthquake: Triggered Flexural-Slip Faulting in the Pamir-Tien Shan Collision Zone

Abstract

Key Points

Plain Language Summary

1 Introduction