Defining the Moho: the Necessity of Combining Different Geophysical Observations

Although the crust was clearly defined as the uppermost layer of the solid earth and has been widely studied by geophysicists, the exact location of Moho---the interface between the crust and the mantle--is still far from clear, at least in some regions. The western part of the North China Craton, namely the Ordos Block, is one such region. In the past ten years, much evidence from receiver-function images has suggested that the Moho beneath Ordos is at ~ 40 km depth. However, a recent study using the newly developed Virtual Depth Seismic Sounding (VDSS) (Yu et al., EPSL, 2012), discovered that the Moho under Ordos is more likely to be at a ~ 60 km depth whereas the interface at ~ 40 km depth is better interpreted as an intra-crustal interface. This raises the question, which seismic method best defines the real Moho? To answer this question, in this study we stress the definition of the Moho as the interface between crustal rock and mantle rock rather than the seismic discontinuity with greatest velocity contrast near the earth surface. Using this re-definition of the Moho, we re-examined the geophysical observations that have been made in Ordos. We found that although the receiver-function method is powerful in imaging crustal and upper-mantle interfaces, it has the shortcoming that it is only sensitive to the S-wave velocity contrast across the interface. Therefore in some regions where the Moho does not have a sharp S-wave velocity contrast, for example in Ordos, the receiver-function image may give the wrong location of the Moho. In such situations, a comprehensive study combining different geophysical observations becomes necessary to determine the Moho. We have shown that VDSS is a very good supplement to the receiver-function technique because it is mainly sensitive to P-wave velocity contrast across the interface. With images provided by both receiver-function analysis and VDSS, we found that the discontinuity at ~60 km depth under Ordos is more likely to be the crust-mantle interface which means that it is the real Moho, although it is rather weak on receiver-function images. In contrast, the interface at ~ 40 km under Ordos, a strong interface on receiver function images, is better explained as an intra-crustal interface because the rock beneath it does not show mantle velocities. This study not only makes the crustal structure of Ordos clearer, but also reminds us that relying merely on one geophysical observation sometimes may give a biased result, so that combining different geophysical observations is necessary to detect the Earth's discontinuities.