Concepedia

Publication | Closed Access

Lateral variations in compressional velocities beneath the Tibetan Plateau from<i>P<sub>n</sub></i>traveltime tomography

76

Citations

27

References

1993

Year

Abstract

A back-projection algorithm is applied to 353 &lt;it&gt;P&lt;inf&gt;n&lt;/inf&gt;&lt;/it&gt; traveltime measurements to image the lateral variations in compressional velocity in the uppermost mantle in and around Tibet. Prior to the tomographic inversion, we refined the apparent &lt;it&gt;p&lt;inf&gt;n&lt;/inf&gt;&lt;/it&gt; slowness measurements based on a recently developed relocation procedure (Zhao &amp; Helmberger 1991), and mathematical formulations of Zhao (1993). These refinements are crucial in obtaining a high-quality velocity image. We obtained an average &lt;it&gt;P&lt;/it&gt;-velocity value for the uppermost mantle of 7.93 km s−1 and an average mantle &lt;it&gt;P&lt;/it&gt;-velocity gradient of 3.1 × 10−3 s−1 for Tibet. The final 2-D &lt;it&gt;P&lt;/it&gt;-velocity image in the uppermost mantle is characterized by a large (7d° to 9d° in dimension) low-velocity (down to about 7.78 km s−1) region in north central Tibet, and by large high-velocity (up to 8.11 km s−1) regions in the western and eastern flanks of Tibet. The crustal thickness is greater than 70 km inside the Tibetan plateau; it decreases toward the southern, eastern and western boundaries of the plateau. At and beyond these boundaries, the crustal thickness is similar to those beneath the Eurasian shield. Rigorous analyses of resolution and error suggest that: &lt;l type="tab"&gt;&lt;li&gt; (1) there is a pronounced difference between the tomographic velocity image obtained with and without the refinements to the &lt;it&gt;P&lt;/it&gt;&lt;inf&gt;n&lt;/inf&gt;-slowness measurements, indicating a significant improvement in the quality of our inversion by these refinements. &lt;/li&gt;&lt;li&gt; (2) Random errors in measuring earthquake origin times and locations, and in the &lt;it&gt;P&lt;/it&gt;&lt;inf&gt;n&lt;/inf&gt; arrival times, should not cause errors of more than 0.1 km s-1 in the velocity image. &lt;/li&gt;&lt;li&gt; (3) The resolution is generally within 6d° to 10d°. &lt;/li&gt;&lt;/l&gt; The velocity image obtained in this study is consistent with previous &lt;it&gt;S&lt;/it&gt;-&lt;it&gt;P&lt;/it&gt; traveltime studies and geothermal/petrological studies, and is consistent with a mantle-convection model in which a convective upwelling is occurring in north central Tibet and downwellings occurring in the surrounding regions (Molnar 1988, 1990).

References

YearCitations

Page 1