Lateral heterogeneity at the base of the mantle revealed by observations of amplitudes of PKP phases

Summary. Several studies have suggested that there are lateral heterogeneities in the velocity structure at or near the base of the mantle. Such heterogeneities can be studied through the analysis of amplitude ratios of core phases PKPAB and PKPDF for given earthquake-station pairs. In the epicentral distance range 155–175° these arrivals are well separated in time, and PKPAB has near-grazing incidence at the mantle—core boundary 80 that it is highly sensitive to lateral changes in velocity structure in that region, while PKPDF with its near-normal incidence is less sensitive. The observed amplitude ratios vary by well over an order of magnitude, but the size of the ratio is found to be correlated with the region of the core—mantle boundary sampled by the PKPAB rays: beneath the eastern Pacific and western Atlantic, for which abundant data are available, well-defined, contiguous regions (millions of square kilometres in extent) are found which correlated with predominantly larger or smaller amplitudes. We interpret this result as indicating differences in the degree to which the velocity structure of the lowermost mantle is heterogeneous.     

Comments on'The palaeomagnetism of the Central Zone of the Lewisian foreland, north-west Scotland'by J. D. A. Piper

Summary. Piper suggested that the Lewisian has rotated 30° anticlockwise since magnetization, whereas the opposite appears more likely. The main magnetization in the Lewisian recognized by Piper and Beckmann was imposed upon cooling after the Laxfordian metamorphism at about 1750 (± 50) Ma. The palaeomagnetic pole corresponding to this magnetization is at 37.6° N, 273.2° E ( dp = 3.7°, dm = 5.2°).
In Greenland, palaeomagnetic poles similar to each other, with a mean pole at 21.6° N, 280.1° E ( K = 52, A ₉₅= 9.4°), have been determined from five widely separated regions in central West Greenland and from Angmags-salik in East Greenland. The magnetization observed in all these regions was established upon cooling after the Nagssugtoqidian metamorphism, again at about 1750 (± 50) Ma.
The Laxfordian and Nagssugtoqidian metamorphisms were equivalent. It is therefore assumed that the two palaeomagnetic poles quoted above were originally identical. Their present difference can be explained by clockwise rotation of north-west Scotland about a local rotation pole since the Lewisian became magnetized, in addition to opening of the Atlantic assuming conventional reconstructions:
(1) assuming the reconstruction of Bullard, Everett & Smith, the local rotation proposed is 39.5° (± 18.1°) about a pole of rotation at 60.3° N, 354.5° E, or
(2) assuming the reconstruction of Le Pichon, Sibuet & Francheteau, the local rotation is 28.0° (±17.7°) about a pole of rotation at 54.1° N, 354.6° E.
These proposals of local clockwise rotation of north-west Scotland accord with that of Storetvedt based on palaeomagnetic results from Devonian rocks on the north-west side of the Great Glen Fault.     

Comment on 'The use of velocity spectrum for stacking receiver functions and imaging upper mantle discontinuities' by H. Gurrola, J. B. Minster and T. Owens

In a recent paper, Gurrola, Minster & Owens (1994; hereafter referred to as GMO) introduced a technique for the analysis of seismic data and applied it to the records of station OBN in Russia. GMO claimed that this was a new technique in earthquake seismology, and that their results for OBN provided a new insight into the deep structure of the Russian platform. We have some comments on their techniques and results.