Limitations of H-κ stacking: ambiguous results caused by crustal layering

详细信息    查看全文

• 作者：I. Wölbern ; G. Rümpker
• 关键词：Teleseismic receiver function ; H ; κ stacking ; Crustal thickness ; Velocity structure
• 刊名：Journal of Seismology
• 出版年：2017
• 出版时间：January 2017
• 年：2017
• 卷：21
• 期：1
• 页码：221-235
• 全文大小：
• 刊物类别：Earth and Environmental Science
• 刊物主题：Geophysics/Geodesy; Structural Geology; Hydrogeology; Geotechnical Engineering & Applied Earth Sciences;
• 出版者：Springer Netherlands
• ISSN：1573-157X
• 卷排序：21

文摘

Over the past decade, the H-κ stacking technique of Zhu and Kanamori (J Geophys Res 105:2969–2980, 2000) has become a standard tool to determine the crustal thickness H and the bulk crustal v_P/v_S ratio κ from teleseismic receiver functions. It is obvious that unfavorable noise conditions as well as a complex 3D velocity structure can severely hamper the interpretation of receiver-function data. However, we observe that ambiguities can even arise from a simple 1D layered velocity structure which raises a high potential for misinterpretations. To analyze the feasibility and basic limitations of the H-κ stacking method, we conduct a series of tests based on synthetic data. The impact of different given elementary parameters, related either to the velocity structure or to the data processing, is evaluated in a series of eight individual tests. We deliberately exclude complications such as 3D structural variations and/or noise to show that even a simple 1D velocity structure, involving, e.g., an additional inter-crustal discontinuity, can have significant consequences for the interpretation of the results. However, our modeling suggests that more complex crustal structures may lead to even less reliable results. Additionally, our tests illustrate that time shifts of the maxima in the H-κ domain due to the superposition and merging of individual phases can lead to significantly overestimated v_P/v_S ratios. In general, the depth to the Moho (or other discontinuities of interest) is less significantly affected. Our tests indicate the necessity to accurately check delay times derived from the maxima of the H-κ stacks against corresponding phases in the receiver functions. Repeating the stacking with varied weighting factors and filter ranges can help to reduce the ambiguities and to avoid possible misinterpretation.