T33F-01
Tectonic Setting of the 8 October 2005 Kashmir Earthquake
The source of the 8 October 2005 earthquake of M 7.6 was the northwest-striking Balakot-Bagh (B-B) fault, which had been mapped by the Geological Survey of Pakistan prior to the earthquake, but, except for a 16-km section near Muzaffarabad, had not been recognized as active. The fault follows the Indus-Kohistan Seismic Zone (IKSZ) and cuts across and locally offsets the Hazara-Kashmir syntaxis defined by the Main Boundary and Panjal thrusts. The fault has no expression in facies of the Miocene-Pleistocene Siwalik Group, but does offset late Pleistocene terrace deposits in Pakistan-administered Jammu-Kashmir. Two en-échelon anticlines near Muzaffarabad and Balakot expose Precambrian Muzaffarabad Limestone and are cut by the B-B fault on their southwest side, suggesting that folding and exposure of Precambrian rocks by erosion accompanied Quaternary displacement along the fault. The B-B fault has reverse separation, northeast side up; uplift of the northeast side accompanies displacement, producing high topography and steeper stream gradients northeast of the fault. Limited surface expression of the B-B fault has been found northwest of the syntaxis, although the IKSZ and steeper stream gradients continue at least as far as the Indus River, the site of the Pattan earthquake of M 6.2 in 1974. To the southeast, northwest-striking faults, one of which is flanked by an anticline exposing Precambrian limestone, were mapped by the Geological Survey of Pakistan. Farther southeast, in Indian-administered territory, active faulting may follow the Riasi thrust, where Holocene activity has been described. In the Kangra re-entrant still farther southeast, active faulting may follow the Soan thrust, along which Holocene and Pleistocene offsets have been described. The Soan thrust, rather than the south flank of the Janauri anticline, may represent the surface projection of the 1905 Kangra earthquake of M 7.8.
T33F-02 INVITED
Large earthquakes in the Himalaya, where, when, how often?
Assessing the characteristics of large earthquakes along the Himalayan arc, such as their rupture extent, magnitude, return period and effects, is a major societal concern as well as a challenging scientific issue. To address that issue a number of complementary approaches need be applied to provide constraints on sub- surface structure, detect active faults and assess their slip rates and seismic behavior. The presentation will first review some aspects of the Mw 7.6 Kashmir earthquake of 2005 as revealed from remote sensing measurements of the surface deformation and seismic waveforms modeling. Next we will review learning made from the study of active deformation in the Himalaya of Nepal, in particular thanks to the combination of morphotectonic studies with geodetic and local seismic monitoring. We will discuss how these data can be used to estimate heterogeneity of geodetic coupling along the Main Himalayan Thrust fault, the relationship to potential large ruptures, implications for the return period of large events and the mechanics governing the observed behavior.
T33F-03
Seasonal variations of seismicity and geodetic strain in the Himalaya induced by surface hydrology as revealed from GPS monitoring, seismic monitoring and GRACE measurements
GPS measurements across the Nepal Himalaya of central Nepal show that long term secular shortening across the range is modulated by seasonal strain variations. Comparison with GRACE measurements of surface load variations and Topex-Poseidon measurements of free surface water elevation of major rivers in the Gangetic plain suggests that the strain variations are probably induced by seasonal variations of surface water storage. Comparison with local seismicity shows that geodetic strain variations also correlate with seasonal variations of seismicity: the seismicity rate is indeed twice as high in the winter as in the summer, and correlates with stress rate variations determined from the modeling of the GPS measurements. This observation provides a way to gauge secular stress build up. We infer ~10-20kPa/yr interseismic stress buildup within the seismicity cluster along the high Himalaya front. Given that Earth tides exert little influence on Himalayan seismicity, the correlated seasonal variation of stress and seismicity indicates that the duration of earthquake nucleation in the Himalaya is of the order of days to month, placing constraints on faults friction laws. The unusual sensitivity of seismicity to small stress changes in the Himalaya might be due to high pore pressure at seismogenic depth.
T33F-04
Pulsed Episodes of Shortening Within the Tian Shan Foreland: Implications for Deformation Rate Interpretations Throughout Central Asia
Unsteadiness in short-term deformation, such as earthquake clustering or development of new faults in lieu of deformation on older structures, poses problems when trying to develop an appropriate context (e.g. regional tectonic rates and active structures) for assessing seismic risk. Here we show that a similar unsteadiness can be found on an orogen scale at million-year time scales, and suggest spatial heterogeneity of deformation rates are a fundamental part of mountain-building in central Asia. The Tian Shan are an intracontinental mountain range that formed north of the Himalayan orogen but in response to the ongoing Indo-Asian collision. Although continuous since the early Miocene, deformation has occurred in an erratic sequence that suggests deformation rates are not constant within the foreland. In order to reconstruct a detailed history of foreland deformation bounding the southern Tian Shan in western China, we have synthesized extensive mapping, analysis of seismic sections, magnetostratigraphy of the foreland fill and associated growth strata, apatite fission-track dating, and changes in sediment-accumulation rates from the >6500 km2 Kashi basin. Three cross-sections spaced along a 55-km-wide, E-W section of the range-front document variable shortening from 11 to 31 km since the initiation of uplift, of which 7-12 km occurred since ~4 Ma. Both overall shortening and total shortening rates throughout the Miocene decrease towards the east, but match the expected differences in magnitude due to 0.7°/M.y clockwise rotation of the Tarim basin around a pole at 93°E, 37°N (Thatcher, W., 2007, JGR 112: B01401). Temporal constraints on individual structures, however, document at least four distinct stages of deformation. Initial uplift (stage 1) of hinterland structures began at 20-25 Ma. Stage 2 occurred at ~16.3 Ma when the basement-involved deformation front stepped south to the Kashi Basin thrust that bounds the foredeep strata. Stage 3 occurred from ~14 to 4 Ma, as the deformation front migrated south episodically above a shallow detachment. Finally, stage 4 began at ~4 Ma with growth of the Keketamu anticline followed by rapid southward migration of the deformation front to the Atushi and then Kashi detachment anticlines. Although average shortening rates since 16.3 Ma were 0.8-1.3 mm/year, rates were faster from 16.3-13.5 Ma (1-3 mm/yr), after which they decreased to 0.5 mm/yr between 13.5 and 4 Ma before accelerating to 2-2.5 mm/yr after 4 Ma. Similarly, the rate of southward propagation of the deformation front was initially ~3-7 mm/yr for the first 10 Ma of basin development, slowed to ~1.5 mm/yr between 13.5 and 4 Ma, and increased to >10 mm/yr since 4 Ma. A mismatch exists between our geologically determined data and present-day geodetic rates: the latter are 3-4 times greater. Furthermore, changes in shortening rates over time suggest that the loci of deformation shifted in and out of the foreland, and imply either changes in tectonic rates or shifting of the locus of tectonic activity throughout the orogen during the Miocene.
T33F-05
Earthquake Surface Rupture of the Salt Range Thrust at the Himalayan Thrust Front in Pakistan
Considerable evidence from Nepal and India now indicates that the basal detachment of the Himalaya produces great earthquakes that result in large coseismic displacements at the thrust front in India and Nepal (the Main Frontal thrust). In contrast, knowledge of the earthquake potential of the Salt Range thrust in Pakistan (SRT) is virtually absent. It has been clear since the publication of the Salt Range maps of Gee (1989) that the SRT deforms young surficial deposits and is an active fault. What remains uncertain is whether surface rupturing events occur on the SRT, with what frequency those events occur, and what is the size of the associated earthquakes. In a field reconnaissance of the SRT in Spring, 2007, we were able to confirm that this thrust is an active fault, and we discovered numerous localities where the fault nearly reaches the surface, cutting all but the youngest few meters of colluvial deposits. Whereas our observations suggest that surface rupturing events occur on the SRT, a number of characteristics of the Pakistani Himalaya suggests the earthquake behavior of the basal detachment and thrust front may be substantially different than it is in India and Nepal to the southeast. Key differences include an uncertain, but lower, convergence rate at the thrust front (5 to 13 mm/yr), a low tapered thrust wedge, and localization of the basal detachment in a weak evaporite unit. In this sense, the front of the Zagros fold-and-thrust belt in Iran may be a more appropriate analog for the thrust front in Pakistan than the Himalayan thrust front to the southeast. Future mapping of deformed geomorphic surfaces and paleoseismic trenching along the SRT will provide the first direct evidence of the earthquake potential and recurrence of plate- boundary earthquakes in Pakistan. This knowledge is critical for hazard assessment in north-central Pakistan where more than 7 million people are likely to be affected by a great earthquake on the plate boundary.
T33F-06
Great paleoearthquakes of the central Himalaya and their implications for seismotectonic models and seismic hazard assessment
A growing body of paleoseismic data collected from more than ten sites in Nepal and India has documented large coseismic displacements at the thrust front (Main Frontal thrust (MFT)). Three great earthquakes have been identified: in ~A.D. 1410 centered north of Delhi, in A.D. 1505 centered in far-western Nepal, and in ~A.D. 1100 centered in eastern Nepal. It is noteworthy that wherever exposures of the MFT have been studied estimates of surface slip are consistently large; with a range of 9-26 m. Historic accounts of the 1505 earthquake describe strong shaking across a 600-km-long stretch of the central Himalaya. A magnitude for this event is estimated to be >Mw 8.5 based on the maximum extent of felt strong shaking, the 100 km width of the locked portion of the basal detachment, and an average slip of 10-15 m. Though no historic accounts exist for the ~1410 and ~1100 earthquakes, the similarity between their surface expression and the 1505 rupture suggests that these events may have been equally large. These surface-rupturing earthquakes are distinctly different from a host of blind thrust events (Mw 7.5-8.4) that dominate the historic record since A.D. 1505. Both blind and emergent earthquakes are presumed to rupture the basal detachment and release interseismic strain that accumulates near the base of the High Himalaya and carry it to the thrust front where Holocene shortening occurs at rates of 15-22 mm/yr. Whereas the surface-rupturing earthquakes clearly deform the thrust front, survey data from the region affected by the 1906 Dehra Dun earthquake suggest that blind events contribute negligible, if any, deformation to the frontal structures. The factors controlling whether or not surface rupture occurs on the MFT remain unconstrained, but the current data seem to suggest that >Mw 8.5 surface-rutpuring earthquakes are the primary contributors to the shortening observed at the thrust front. It is sobering to consider that the ‘Big One' has not struck the Himalaya in over 500 years and that Mw 7.5-8.4 earthquakes are the ‘moderate' earthquakes'. Further study to constrain the lateral extent and recurrence of the great paleoearthquakes of the central Himalaya is critical to answer important questions about the Himalaya earthquake cycle and the seismic hazard facing the rapidly urbanizing population of the region.
T33F-07
October 08, 2005 Muzaffarabad Earthquake: New Data on the Indus Kohistan Seismic Zone and its extension into the Hazara-Kashmir Syntaxis, NW Himalayas of Pakistan
This paper deals with the data obtained from local networks in northern Pakistan for 251 earthquakes of magnitude ¡Ý 4.0 for October 8, 2005 to December 31, 2006 period. The study presents focal mechanism solutions (FMS) of 12 pre- (1904-2005) and 17 post- (October 8, 2005-December, 2005) Muzaffarabad Earthquake, their detailed tectonic interpretation, and correlation with surface evidence of co-seismic rupture with published Synthetic Aperture Radar (SAR) data. Distribution of landslides obtained from National Engineering Services of Pakistan and the earthquake damages are also discussed. Aftershock distribution, which is more prominent in the crystalline zone (northwest of Muzaffarabad), defines a 50 km wide NW-SE trending zone that extends for 200 km from the Main Mantle Thrust to the centre of the Hazara Kashmir Syntaxis (HKS). The FMS of the main shock and 16 aftershocks having magnitude ¡Ý 4.0 indicate thrusting to be the dominant mechanism with rupture planes having NW-SE trend and NE dip. In addition, 12 FMS of pre-Muzaffarabad Earthquake (1904- 2004) from the same area have been determined and results are compared. This leads to the conclusion that the wedge-shaped NW-SE trending blind zone, referred to by earlier workers as the Indus Kohistan Seismic Zone (IKSZ), has been activated during the Muzaffarabad Earthquake. The right-lateral component in all FMS, supported by the surface evidences, suggests the involvement of Balakot-Bagh Fault (BBF). We propose that the IKSZ is the source of the October 8, 2005 Muzaffarabad Earthquake that reactivated the BBF. Furthermore, the IKSZ does not end at the nose of the syntaxis, but extends further southeast of it. Tectonic complexity seems to be due to a variety of factors. Also, thrust and reverse solutions near the northern collisional boundary (Main Mantle Thrust) have mostly NE-SW directed P-axis orientations. From the detailed FMS analysis, three conclusions have been drawn: 1) Shallow events (depth ¡Ü 10 km) with prominent strike slip solutions (earlier earthquakes) are associated with the surface strike- slip faults (e.g., Muzaffarabad Fault) and/or the Besham domal structure, 2) Moderate depth events (depth 10-25 km) with thrust/reverse solutions but having minor right-lateral strike slip component (all Muzaffarabad earthquakes and two earlier) are associated with the IKSZ, 3) Deeper earthquakes (depth below IKSZ) with pure thrust/reverse solutions may be related to the under-thrusting of the Indian plate beneath the IKSZ, which represents a major thrust zone. Imbricate thrusting, and breaking and thickening of the crust are considered to be caused by steep bending of the under-thrusting plate at the collisional boundary.However more seismological data is required to prove these hypotheses.