Graduate Thesis Or Dissertation
 

Investigating the Tibetan crust through automatic S wave detection and travel-time tomography using the Hi-CLIMB seismic array

Public Deposited

Downloadable Content

Download PDF
https://ir.library.oregonstate.edu/concern/graduate_thesis_or_dissertations/12579w56f

Descriptions

Attribute NameValues
Creator
Abstract
  • The Hi-CLIMB broadband seismic experiment (2002-2005) operated 233 stations along an 800 km long north-south line from the Himalayan foreland into the central Tibetan Plateau and in a 350x350 km sub-array within southern Tibet and central and eastern Nepal. Station spacing was approximately 8 km along the line and 50 km within the sub-array. The experiment spanned the Himalayan range, Lhasa Block, Qiangtang Block and crossed the Yarlung Tsangpo Suture (YTS) and the Banggong-Nujiang Suture (BNS). From June 2004 to August 2005, over 22,500 local and regional seismic events were recorded throughout the south-central Tibetan Plateau based on automated arrival time picks. This dataset provides an opportunity to jointly invert for crust and upper mantle velocity structure along with earthquake locations using both P and S waves. Automated P and S wave picks, however, were originally determined from vertical component data using short term average to long term average (STA/LTA) windows, resulting in relatively few S picks of generally low quality. To increase the number of accurate S arrivals, we implemented an automatic S-wave picker, which uses signal attributes from three-component seismic data. The signal attributes used are rectilinearity, directivity relative to incoming P wave, ratio of transverse to overall energy and transverse amplitude. An S pick is declared when the combination of signal attributes reaches a noise dependent threshold. We used manual picks and S phase observations from events throughout south-central Tibet to adjust picking parameters and thresholds to optimize automatic S picks. For shallow events we found S[subscript g] can be picked reliably to the S[subscript g]/S[subscript n] crossover distance of approximately 3° while S[subscript n] arrivals are absent. Deep events beneath the southern Tibetan Plateau and the High Himalayas produce clear S arrivals that can be picked to about 7° distance. Applying the S-picker to over 5,700 larger (M[subscript L]≥2.7), well-recorded events led to about 100,000 S picks, significantly increasing the number of arrivals and improving their accuracy. Compared to STA/LTA picks, the new automatic S picks show a decrease in average arrival time residual by over 30 percent. This new polarization picking process allowed us to increase ray coverage, which is crucial for P and S wave structural inversions. With the new automatic arrival time picks we use tomoDD double-difference tomography to invert for crust and uppermost mantle structure beneath the Tibetan Plateau. We use the same subset of over 5,700 best-recorded local events containing over 200,000 P and 100,000 S wave arrivals to conduct the inversion. In the upper crust we observe extensive low velocity zones extending to 25 km depth. Based on Vp and Vp/Vs results we suggest this material is quartzite and highly felsic granite. Beneath the BNS we image low Vp values and increased seismicity, indicating a possible fault zone down to 25 km depth. In the lower crust beneath the Qiangtang Block we image high Vp and increased seismicity, suggesting an area of increased crustal strength to the north. Beneath the High Himalaya, south of YTS, the Moho depth decreases from 55 to 75 km depth as the Indian plate subducts beneath Tibetan crust. Along the Moho we observe Vp values upwards of 8.5 km/s extending to 31°N. We attribute these high velocities to the formation of eclogite along the base of the Indian plate, which terminates at 31°N as upper mantle velocities drop to 8.2 km/s. In the lower crust beneath the Lhasa Block low Vp values, around 6.5 km/s, extend to near Moho depth. A lack of seismicity is also present, which could indicate a weak and possibly ductile lower crust in the Lhasa Block. Nowhere in the lower crust does the Vp/Vs ratio exceed 1.8 suggesting extensive areas of partial melt are unlikely.
License
Resource Type
Date Available
Date Issued
Degree Level
Degree Name
Degree Field
Degree Grantor
Commencement Year
Advisor
Committee Member
Academic Affiliation
Non-Academic Affiliation
Subject
Rights Statement
Publisher
Peer Reviewed
Language
Replaces

Relationships

Parents:

This work has no parents.

In Collection:

Items