UPPER-MANTLE ANISOTROPY BENEATH CENTRAL-EUROPE FROM SKS WAVE SPLITTING - EFFECTS OF ABSOLUTE PLATE MOTION AND LITHOSPHERE-ASTHENOSPHERE BOUNDARY TOPOGRAPHY

Fast polarization directions cu of split SKS waves in Central Europe c hange from NE/ENE in the western part to dominatingly E/ESE orientatio n towards north and east. This coincides strikingly well with the domi nating trend of Hercynian deformational crustal features, It hints to frozen anisotropy related to paleo-crustal fabric. But when considerin g plausible anisotropy values of about 2-3% then only a small fraction (delta t < 0.3 s) of the rather large observed average delay-times (d elta t = 0.83 +/- 0.31 s) between the two split waves could be attribu ted to structural anisotropy in the relatively thin Central European c rust. Therefore, the main ''anisotropy signal'' has to be associated w ith lattice-preferred orientation (LPO) of olivine below the crust, It may be either frozen in the subcrustal lithosphere since Hercynian ti mes or have developed more recently in the asthenosphere. The thicknes s of the lithosphere varies significantly beneath Europe and the depth contours show systematic changes in trend, The latter varies from dom inatingly NE in the southwest to SE in the north and east. The polariz ation directions a of the fast split SKS waves observed at seismic sta tions in proximity to the southern and northeastern boundaries of Cent ral Europe are subparallel to the trends of these strong anomalies in lithosphere topography. A causal relationship is assumed and a new mod el proposed to explain the observations in alpha and delta t. It takes into account the possible effects of paleo-deformational events. They may have produced both anisotropic crustal fabric and probably still preserved and similarly trending frozen LPO in the subcrustal lithosph ere. The model also considers the influence of recent absolute motion of the West European lithospheric plate towards NE and the effect of i ts pronounced lower boundary topography on the formation and trend of LPO in the asthenosphere. Accordingly the effects of anisotropy of dif ferent nature and age at different depth levels but with similar trend may superimpose constructively. This could explain the rather large d elay-times observed at Central European stations which are too large t o be attributed to frozen anisotropy in the lithosphere alone. The mod el would even permit the total effect observed to be attributed to ast henosphere flow controlled by absolute plate motion direction and lith osphere-asthenosphere boundary topography. Copyright (C) 1996 Elsevier Science Ltd