Search

Body waves are routinely observed in cross-correlations of ambient noise records from seismic arrays worldwide but it is still poorly understood how the lack of uniform coverage of body wave sources at sea impacts reconstruction of some deep seismic phases such as, for example, reflections on mantle discontinuities. These difficulties presently hinder a generalization of the extraction of these phases, even though the feasibility of such techniques has been demonstrated previously. In this study, we focus on the extraction of P reflections from the 410 discontinuity (Pv410P) using cross-correlations from records of a quasi-linear array between Tibet and the Himalayas. We demonstrate that in this geographic location and with the network geometry at hand, the noise correlations in the second microseismic peak are during most of the year dominated by a noise source in the north Pacific. The potential Pv410P phase revealed in the stacks of all data and through singular value decomposition analysis on a subset of data has an amplitude that is similar to that of cross terms of different body wave phases from the source area, so a positive identification of the Pv410P remains speculative. As the arrival times and apparent velocity of mantle-reflected waves are close to those of several cross terms, other diagnostic tools are needed to identify areas and network configurations for which noise correlations can confidently be used to analyse reflections from the mid-mantle discontinuities. [ABSTRACT FROM AUTHOR]