ALBERT

Description: Globally, one of the largest intraplate earthquakes of M w  7.7 occurred on 26 January 2001 in the Kachchh rift basin (KRB), western India. The continuing long aftershock sequence over decades has generated much debate on the seismogenic fault(s). We have analyzed more than 10,000 aftershocks ( M w 〉1.0) recorded by a 50-station broadband network in the region during 2006–2014. A total of 834 aftershocks ( M w 〉2.4), each recorded by at least eight broadband seismic stations with a minimum of eight P and six S phases, are relocated in this study by double-difference tomography (tomoDD) method. The relocated aftershocks and velocity images reveal a near-vertical or steeply south-dipping deeper structure as the source zone of the mainshock and aftershocks; the structure correlates well with the geologically mapped South Wagad fault (SWF). Among the other geologically known faults, the Kachchh Mainland fault (KMF) and the Gedi fault (GF) are also well identified in the seismic sections. Further, fault-plane solutions of 109 aftershocks having M w ≥3.5 corroborate well with the known faults. The geological model and seismological observations suggest that the SWF overstepped the KMF and intersects it at depth. The intersecting fault zone is the source area for the deeper (10–35 km) reverse faulting earthquakes in the KRB. At the fault end of the SWF, a heterogeneous velocity structure is imaged, which is attributed to a fluid-filled rock matrix that triggered the mainshock. On the other hand, the GF is reported to be a later-activated fault to the north of SWF; it generated some shallower aftershocks (0–20 km) mostly by strike-slip mechanisms.

Description: An attempt is made to map the spatial variation of the tectonic stress pattern in northeast India and its adjoining south Asia region using stress tensor inversion of some 516 fault-plane solutions. The Bhutan Himalaya and the Arunachal Himalaya are mapped with north–south to north-northwest–south-southeast compression. The eastern Himalaya syntaxis zone, on the other hand, shows a clockwise rotation; a north-northeast compression is dominant. To the south, in the intraplate part of the region, the Shillong plateau, Assam valley, Bengal basin (Bangladesh), and Tripura fold belt exhibit north-northwest to north-northeast compression. Orthogonal horizontal extension is dominant in southern Tibet, Bhutan, and partly in the syntaxis zone, and the same is also observed in the Shillong plateau and Assam valley area of the intraplate region. The Indo–Burma ranges and the Sagaing fault in the Myanmar region show a northeast–southwest compression; an orthogonal horizontal northwest–southeast extension is also observed in the Sagaing fault zone. A depth variation of the tectonic stress is observed below the Indo–Burma ranges; it changes from north–south to northeast–southwest in the southern part, and from northeast–southwest to north-northeast–south-southwest in the northern part in the deeper seismogenic zone. The stress inversion results of clusters of events in individual zones, though mostly conformable with the average observations, indicate a variation in the Shillong plateau due to heterogeneity and tectonic complexity.

Description: We estimated frequency-dependent attenuation of coda waves ([IMG]/medium/109eq1.gif" ALT="Formula "〉) and body waves ([IMG]/medium/109eq2.gif" ALT="Formula "〉 and [IMG]/medium/109eq3.gif" ALT="Formula "〉) in 1.5-24 Hz by applying the single isotropic scattering theory and the extended coda-normalization method, respectively, in the crust beneath the Andaman Sea. We used 43 aftershocks of the 13 September 2002 earthquake (Mw 6.5) in the Andaman Sea recorded by three stations installed in the Andaman Islands. The coda Q factors calculated from the amplitude decay rate of the S-wave coda show a dependence on frequency and lapse time. We found that with the increase in lapse time window from 10 to 40 s, Q0 (QC at 1 Hz) increases from 55 to 153, while the frequency-dependent coefficient n decreases from 1.1 to 0.94. The average frequency-dependent relations of [IMG]/medium/109eq4.gif" ALT="Formula "〉 vary from 0.02f-1.1 to 0.01f-0.94 with an increase in lapse time window from 10 s to 40 s, respectively. The values of [IMG]/medium/109eq5.gif" ALT="Formula "〉 and [IMG]/medium/109eq6.gif" ALT="Formula "〉 corresponding to spectral amplitude decays show strong frequency dependence and are expressed as 0.02f-1.01 and 0.01f-1.0, respectively. Our results are consistent with those of other seismically active regions. The ratio [IMG]/medium/109eq7.gif" ALT="Formula "〉 is found to be larger than unity for the whole frequency range. We separated intrinsic absorption ([IMG]/medium/109eq8.gif" ALT="Formula "〉) and scattering attenuation ([IMG]/medium/109eq9.gif" ALT="Formula "〉) using the independent estimates of [IMG]/medium/109eq10.gif" ALT="Formula "〉 and [IMG]/medium/109eq11.gif" ALT="Formula "〉. The results show that [IMG]/medium/109eq12.gif" ALT="Formula "〉 is close to [IMG]/medium/109eq13.gif" ALT="Formula "〉 and both of them are larger than [IMG]/medium/109eq14.gif" ALT="Formula "〉, suggesting that coda decay is predominantly caused by intrinsic attenuation.