SPICE-Research Training Network - UiO > Polarisation anomalies > The Kerguelen hotspot
anisotropy on the Kerguelen hotspot track inferred
from Rayleigh wave polarisation anomalies.
Page 3 of 6
Page 1 | 2 | 3 | 4 | 5 | 6
Wave path deviations from the great circle cannot explain the polarisation anomalies:
Wave paths may deviate from the epicentre-receiver great circles, what may result in Rayleigh waves polarisation anomalies. We quantify this possible contribution for this dataset: the maximum deviation of 5.3° was obtained resulting in a transverse component equal to 9 per cent of the longitudinal one - smaller than the 15 per cent detection threshold.
Although they are far from negligible for the most distant events, the wave path deviations remain small compared to the polarisation anomalies we observe. This is partly related to the fact that for the relatively short epicentral distances used here, wave path deviations remain small in tomographic models of the Earth. Even when taking into account that phase velocity contrasts may be underestimated in tomographic models, deviation related to large-scale phase velocity variations can only account for a small fraction of the polarisation anomalies.
Wavefield distortion by the local structure of the Kerguelen Plateau: isotropic and anistropic models.
Fig. 4: Sea floor topography of the Kerguelen Plateau. The black lines indicate the boundaries between the oceanic basins, the northern plateau and the central plateau in our models, based on those in Shipboard Scientific Party (2000). The digital bathymetry data are derived from Geosat and ERS-1 gravity information (Smith & Sandwell 1997). (after Pettersen and Maupin, Fig. 13, GJI, 149 225-246)
We analyse here if the structure of the Kerguelen Plateau, which is quite different from the normal surrounding oceanic basin, can distort significantly the elastic wavefield and cause the observed polarisation anomalies.
Structure of the Kerguelen Plateau:
Following Könnecke et al. (1998), the Kerguelen Plateau can be divided into four domains: the northern, central and southern plateaux, and the Elan banks. The southern and central plateaux and Broken Ridge, the latter now being located ~1800 km north-east of the Kerguelen Plateau, were formed during the Cretaceous by voluminous outpouring of magma associated with the Kerguelen hotspot.
The edge of the plateau, as defined in Shipboard Scientific Party (2000), has a complex shape, especially to the North-East (see figure above). It is surrounded by oceanic basins with age ranging from 40 My to the North-East to 80 My to the South-East (Schlich 1982). Müller et al. (1993) proposed that the hotspot is presently located beneath the western part of the northern plateau.
The crustal structure of the Kerguelen Plateau has been investigated by a series of seismic refraction profiling surveys (Charvis et al. 1995; Operto & Charvis 1996; Charvis & Operto 1999). The northernmost surveys have been done on the isles themselves, where the crust was found to be 16-20 km thick, which is probably representative of the structure of the whole northern plateau. A profile to the south-east of the isles, on the central plateau, shows a different type of crust: a thickness of 24 km and a particularly thick lower crust, consistent with an Icelandic-type of crust (on-ridge setting in the presence of hotspot).
We have no direct information on the seismic structure deeper in the mantle under the Kerguelen Plateau. However, models for similar plateaux exist. The mantle under Broken Ridge shows no anomaly compared to the oceanic mantle around, and the mantle under the Ninetyeast Ridge shows low velocities of 4.5 km/s down to about 45 km depth. Under the Ontong Java Plateau, low velocities extend much deeper, down to at least 150 km depth. The crust is also much thicker than at the Kerguelen Plateau, with an average Moho depth of 33 km.
Based on these informations, we build a 3-D model of the Plateau.
Page 3 of 6
Page 1 | 2 | 3 | 4 | 5 | 6
Back to Top | Previous | Contact | Home
© Universitetet i Oslo - Institutt for geofag / SPICE 2004