No abstract

Magnetotelluric (MT) and geomagnetic depth sounding (GDS) data were recorded at 25 locations across Vancouver Island, Canada, over the subducting Juan de Fuca plate. The GDS data respond to a number of local conductive features at frequencies above 10 Hz while at longer periods, the in-phase and quadrature induction arrows axe colineax and are aligned approximately normal to the continental shelf with a maximum amplitude response neax the southwest coast. Interpretation of the MT data show the presence of a conducting zone at depths greater than 20 km beneath Vancouver Island and correlating with a strong seismic reflector. Extensive testing with two dimensional models shows the MT and GDS data axe consistent with a proposed geoelectrical structure containing a conducting layer extending from the Pacific Ocean, dipping beneath Vancouver Island, and in electrical contact with the mainland conductive region. The layer has a horizontal conductance of 200 S that may decrease to 100 S beneath the northeast side of Vancouver Island. Related geophysical studies suggest the conductive zone is above the actual tlmast surface of the Juan de Fuca plate. The conducting layer is most likely the result of cracks and pores filled with saline fluids which axe supplied by water subducted with the oceanic crust and by dehydration reactions. The presence of fluids has significant implications for thrust eaxthquakes and for metamorphic reactions that occur in subduction zones. The conductance of the upper mantle (between 90 and 390 km) is best modelled at approximately 1500 S. of the plate provides an excellent target for studying part of the western North American plate which is curthe electrical structure of a subduction zone. Kurtz et rently overriding the Juan de Fuca plate. Much of Vanal. [1986] presented a summary of the MT interpre-couver Island is part of the Wrangellia terrane [Jones et al., 1977], a thick assemblage of Paleozoic, Mesotation results including a highly conducting zone near the top of the subducting plate at the same depth as a zoic and Cenozoic volcanic, plutonic, sedimentary and strong seismic reflective zone [Green et al., 1986]. This metamorphic rocks [Muller, 1977]. Wrangellia was acpaper gives a more detailed interpretation and includes expanded discussions of the MT and GDS data.

Magnetotelluric sounding reveals a conductive zone beneath the Pacific Ocean at depths in excess of 60 km but does not resolve the resistivity of the lithosphere above this zone. Further resolution can be obtained by controlled source electrical methods. The simplest of these are the galvanic techniques. Dipole‐dipole resistivity sounding is not suitable because dipole separations of thousands of kilometers would be required to obtain values of the resistivity. A viable alternative is to measure on the ocean floor the magnetic field of a vertical bipolar source extending from the sea surface to the seafloor. Magnetometer transmitter separations of only a few kilometers are sufficient to determine the resistivity of a half space beneath the ocean. Sounding curves similar to those of the resistivity method may be constructed to resolve the resistivity of a layered lithosphere. The curves constructed are valid at alternating frequencies which are small compared with a skin frequency not in the ocean but in the lithosphere. The depth of penetration is of the order of half the transmitter‐receiver separation. Magnetic field amplitudes are in the range of picotesla for reasonable lithospheric resistivities and separations up to 10 times the length of the bipole. Modern instrumentation, modified for the ocean floor, can detect such signals at a range of 20 km at a frequency of about 0.02 Hz averaged over several hours.

Magnetic variations measured a t three stations across the strike of the anomaly in electrical conductivitv of the earth's mantle suerested bv Whitham and Andersen (1962) near Alert, Elles~nere Island, havelbeen analyzed by simple potential theory. T h e anomalous internal contributions to the magnetic variation 1-ector show striking confinement consistent with an underground current in a northeast-southwest direction a t a depth between 50 and 70 km, a few kilometers southeast of Alert. If the anomalous c o n d~~c t o r is approxin~ated by a uniform infinite cylinder, the horizontal field response as a function of frequency can then be explained with a conductivity of ~3 X 1 0 -'~ e.m.u. and a cylindrical radius of about 50 km. T h e vertical field response is, however, an unsatisfactory fit to this model. A possible asymmetry is noted based on the hourly range data: this could be explained by the dipping of the real conductor under Ellesmere T~l a n r lPreliminary earth potential measurements show that, a s expected, the electric field variations are abnormally low a t Alert.The gravitational consequences of the cylindrical model are discussed and compared with two profiles from the same region. A Bougucr anomaly is found in the predicted region if certain regional gradients are assumed, but its magnitude is snlaller than that predicted from the cylindrical model in hydrostatic equilibriunl. I t is clear that approximate agreement can be obtained with crustal thinning of some 20 km.The significance of the thermal ano~naly thought t o be responsible for the electrical conductivity anomaly is discussed briefly.

No abstract