Possibility of a low P-wave velocity layer in the outermost core from global SmKS waveforms

Possibility of a low P-wave velocity layer in the outermost core from global SmKS waveforms
复制标题

DOI:
10.1016/j.epsl.2007.05.007
复制
发表时间:
2007-07
影响因子:
5.3
通讯作者:
S. Tanaka
S. Tanaka
中科院分区:
地球科学1区
文献类型:
--
作者:
S. Tanaka

文献摘要

相似文献

利用全球1211个SmKS(m≥2)波形资料,研究了核幔边界附近的径向地震速度结构。带通(f=0.02-0.1 Hz)和叠加波形与PREM的反射率合成波形吻合得很好,而其他全球模型(iasp 91,ak 135和SP 6)的波形则不一致。以PREM为初始结构的最外层岩芯的初步波形模拟结果表明,岩芯表层在岩芯顶部的P波速度为7.95 km/s,厚度为90 km。基于三维全球S波速度模型,本研究中收集的SmKS相位主要穿透地幔底部的轻微高速区域,预计S3 KS-SKKS的差分走时残差为负值,平均值约为-0.5 s。然而,正S3 KS-SKKS残差(平均值为+0.3 s)在叠加波形中占主导地位。此外,对D″构造的波形模拟可以得到30 km厚的地幔底部(ULVZ)S波速度降低10%的层,其中主要S3 KS部分的波形拟合得到改善。另一方面,几百公里厚的地幔最下部构造对波形资料解释不好。这表明SmKS数据仍然受到地幔最低几十公里的结构的影响,而全球层析成像没有充分模拟。虽然薄的低S波速度模型是不确定的,低P波速度层在最外面的核心的可能性仍然是因为波形适合的S4 KS的一部分,改善了ULVZ和140公里厚的层与0.8%的P波速度降低在核心顶部。
A global data set consisting of 1211 SmKS (m≥2) waveforms has been analyzed to investigate the radial seismic velocity structure around the core–mantle boundary (CMB). Band-passed (f=0.02–0.1 Hz) and stacked waveforms coincide with reflectivity synthetic ones for PREM very well, whereas those for other global models (iasp91, ak135, and SP6) yield disagreements. A preliminary waveform modeling for the outermost core with PREM as initial structure results in a core surface layer with a P-wave velocity of 7.95 km/s at the core top and thickness of 90 km. Based on a 3-D global S-wave velocity model, SmKS phases collected in this study mainly pierce slightly high velocity regions at the base of the mantle, and the predicted differential travel time residuals for S3KS-SKKS are expected to be negative, approximately −0.5 s for the averages. However, positive S3KS-SKKS residuals, of which average is +0.3 s, are predominant in the stacked waveforms. Moreover, a waveform modeling for the D″ structure can result in a 30 km thick layer with a 10% S-wave velocity reduction at the mantle bottom (ULVZ), in which waveform fitness for the part of mainly S3KS is improved. On the other hand, the waveform data is not well explained by the lowermost mantle structures with thickness of several hundred kilometers. These suggest that the SmKS data is still affected by the structure of the lowermost several tens kilometers in the mantle that is not sufficiently modeled by a global tomography. Although the thin low S-wave velocity model is not conclusive, the possibility of a low P-wave velocity layer in the outermost core is remained because the waveform fitness for the part of S4KS is improved by the combination of the ULVZ and a 140 km thick layer with a 0.8% P-wave velocity reduction at the core top.