Applied and Fundamental EM and CSMT in Ocean Studies Chairpersons: R.L. Evans and C. Brown 2.1 EM FIELDS INDUCED BY A REALISTIC MODEL OF THE OCEAN TIDE Agusta H. Flosadottir(1) and G.D. Egbert(2) (1) HALO Inc. Reykjavik, Iceland and University of Washington, USA (2) COAS, Oregon State University, Corvallis, OR, USA agusta@pmel.noaa.gov In its sweep across the Earth's magnetic field, the ocean tide generates strong electromagnetic signals. Such signals have been observed on the seafloor with electrometers and undersea cables in a variety of locations, and have even been seen far inland on the continents. One of the outstanding questions relating to these observations is the sensitivity of tidally generated fields to the electric conductivity structure of the underlying and surrounding Earth, and whether there are circumstances where this could be exploited for geophysical sounding. Another important question relates to the spatial scales of electric current loops generated by the tide, and the importance of self- and mutual induction in the interpretation of the signals. To approach these questions, currents from a numerical representation of the ocean tide are used to generate sources to drive a 3-d model of the generated electromagnetic fields. The tidal currents are derived by assimilating TOPEX/POSEIDON satellite altimeter data into a linearized barotropic model of tidal dynamics. The electromagnetic model is based on the 3-D magnetotelluric code of Mackie and Madden, modified to permit induction sources within the conducting ocean. 2.2 STUDIES OF CURRENT VARIABILITY IN THE SEA OF JAPAN USING LONG-TERM CABLE VOLTAGE MEASUREMENTS N.A. Palshin(1), L.L. Vanyan(1), R.D. Medzhitov(1), G.I. Shapiro(1), M.A. Evdoshenko(1), H. Utada(2), H. Shimizu(2), and Y.Tanaka(3) (1) Shirshov Institute of Oceanology, Moscow, Russia (2) Earthquake Research Institute, University of Tokyo, Japan (3) Faculty of Sciences, Kyoto University, Japan palshin@geo.sio.rssi.ru By the present JASC cable voltages time series since April 1996 till December 1999 with a sampling rate 1 Hz have been collected. Seasonal variability of water transport across the central part of the Sea of Japan has been revealed as a result of statistical processing. Data processing of daily mean values revealed the dominance of the motionally induced signals in the measured voltages for the periods greater than several days. The analysis of the data gives possibility to separate the short-period fluctuations with the period about 10 days. The particular emphasis has been placed on studying the origin of ten-diurnal variation. It was proposed that this phenomenon could be explained with several main sources: the meso-scale eddies passing through the Sea of Japan, temporal instability of the main current paths and sea interaction with the atmosphere. The numerical modeling results validated the possibility of explaining of cable voltage fluctuations with period about 10 day by the meso-scale eddies moving along the slope across the cable. The joint analysis of the hydrometeorological data and long-term voltage measurements has shown that there are time intervals, within which the correlation between the processes under study exists. 2.3 ASYMMETRIC MANTLE ELECTRICAL STRUCTURE BENEATH THE EAST PACIFIC RISE AT 17S Alan D. Chave(1), Rob L. Evans(1), Pascal Tarits(2), and MELT Team (1) Woods Hole Oceanographic Institution, Woods Hole, MA 02543, USA (2) Universite Bretagne Occidentale, Brest, France alan@whoi.edu The magnetotelluric component of the Mantle ELectromagnetic and Tomography or MELT experiment measured the electrical conductivity structure of the mantle beneath the fast-spreading southern East Pacific Rise. The most prominent feature of the measurements is a highly asymmetric electrical structure, with higher conductivity to the west of the spreading ridge. The uppermost 100 km of mantle immediately to the east of the ridge crest is consistent with dry olivine, suggesting a mantle depleted of melt and volatiles. Mantle conductivity to the west of the ridge is consistent with a low fraction of about 1-2 percent interconnected melt distributed over a broad region and extending to a depth of about 150 km. The asymmetry in electrical structure may be the result of asymmetric spreading and westward migration of the ridge axis, and suggests distinct styles of melt formation and delivery in the mantle beneath the two plates. Other features of the models will be tested for significance and discussed in detail. 2.4 CORRECTING THE MT DATA FOR TOPOGRAPHIC EFFECTS TO IMAGE THE SHALLOW MELT UPPER MANTLE STRUCTURES Marion Jegen and Pascal Tarits IUEM, Universite de Bretagne Occ., France jegen@univ-brest.fr The understanding of the flow of solid and molten material at mid- ocean ridges is hampered by the lack of knowledge of vital physical parameters. These are, for instance, the amount, distribution and connectivity of partial melt and the temperature distribution in the subsurface. Changes in the geometry of the distribution of melt as well as the amount of melt present in the solid material have a large impact on the electrical conductivity, it changes the electrical conductivity by orders of magnitude. The sensitivity of the electrical conductivity to partial melt has led to the MELT-EM experiment. The MELT data consists of MT data, i.e. measurements of the electrical impedance as a function of frequency at 36 sites on the East Pacific Rise. The data acquisition itself has been completed and an initial model of a 2D conductivity distribution in the subsurface has been derived. During the interpretation of the MT data it became quickly clear, that the topography of the seafoor could severely bias the data at the higher frequencies above 5 10-3 Hz, impeding imaging the shallowest structures at a depth of less than 20 km beneath the ridge crest. Due to the high conductivity contrast between the ocean and the seafloor, the topography deforms the induced currents along the seafloor where the data is measured. Previous approaches allowed to correct for this bias at low frequencies impedance measurements. These low frequencies contain information of the electrical conductivity at great depths below 30 km and the correction therefore yielded the determination of the conductivity contribution in the initial model at depth. For high frequency data, which contain information about the upper 20 km, this scheme is insufficient. However, this region is especially interesting since most of the melt is contained in this region. It is therefore important to address the problems introduced by topography at high frequencies. We are presenting a model study that shows how the topography influences the measured data at high frequencies in detail. Based on this model study we have derived a scheme with which the topography effect can be taken into account at these frequencies and show how this correction changes and improves the interpretation the high frequency data in the initial model. 2.5 CONDUCTIVITY STRUCTURE OF A MID-OCEAN RIDGE: MT RESULTS FROM THE EAST PACIFIC RISE Kerry W. Key and Steven C. Constable Scripps Institution of Oceanography, USA kkey@ucsd.edu In spite of recent intense geophysical examination of ridges, we still know surprisingly little about the spatial extent and temporal evolution of melt migration and storage. Seismic techniques can image the top of crustal magma chambers but are poorer at describing the distribution, vertical extent and volume fraction of melt. EM methods are sensitive to total connected melt, but have proven difficult in the past: controlled source methods have limited depth penetration and marine MT instruments were limited to long period signals which sampled deeper mantle conductivity. Recently Scripps Institution of Oceanography has developed equipment which expands the high frequency range of marine MT data collection, and we now routinely collect 1 Hz data in 1 km of seawater. A pilot study conducted at the fast-spreading East Pacific Rise at 9 50 north consisted of 8 seafloor MT sites sampling at 25 Hz for three weeks. At the time of writing the data set is only a few days old but preliminary robust multistation array processing shows good MT responses from 10 seconds to several thousands seconds period. Seismic studies and bathymetry data suggest a dominantly 2-D structure. The MT responses are indeed strongly 2-D and aligned with the ridge axis. Some of this signal is probably due to topography but 1-D models of the TE mode suggest a highly conductive body at a depth of a few kilometers. We will present the complete data set and 2-D forward and inverse model studies at the meeting. 2.6 THE RESISTIVITY STRUCTURE OF CONTINENTAL SHELF SEDIMENTS: SURVEY ON THE EAST AND WEST COAST U.S. MARGINS Rob L. Evans(1), Lawrie Law(2), and B. St Louis(3) (1) Department of Geology and Geophysics, WHOI, Woods Hole, MA 02543, USA (2) Pacific Geoscience Centre (retired), North Saanich, BC V8L 4B2, Canada (3) Geological Survey of Canada, Geophysics Division, Ottawa, Ontario, Canada revans@whoi.edu Two large electromagnetic (EM) data sets have been collected in different sedimentary environments, one off California and the other off New Jersey. The data were collected using a towed EM surveying system owned and operated by the Geological Survey of Canada. The system uses frequency domain magnetic fields to measure the electrical resistivity of the uppermost 20 m or so of the seafloor. The survey areas were targeted by the Office of Naval Research as focus areas for the STRATAFORM initiative. Because of this, there is a wealth of additional and coincident data in both regions, allowing us to compare EM data with other techniques, such as coring, sidescan sonar and high resolution bathymetric mapping. Highlights of the data include a region of anomalously high resistivity off California, that is closely related to a shallow anticline system and which may represent a region where fresh groundwater is migrating to the seafloor. Off New Jersey, we see the clear signature of buried paleo-channels that were previously seismically imaged. We will present details of the data acquisition and interpretation and discuss future directions for the technique. 2.7p MEASUREMENTS OF THE ELECTRIC FIELD INDUCED BY SEA CURRENTS IN THE COASTAL ZONE OF THE STRAIT OF WHITE SEA N.A.Palshin(1), V. A. Matyushenko(2), L.L. Vanyan(1), and A.M. Poray-Koshits(1) (1) Shirshov Institute of Oceanology, Moscow, Russia (2) Institute of Ecological Problems of the North Arkhangelsk, Russia palshin@geo.sio.rssi.ru Low-frequency electromagnetic field in seas and oceans provide information on the variability of the water currents. Measurements carried out in the Strait of the White Sea, northwestern Russia in June 1998 and August 1999 were aimed to studying the possibility of detecting motionally induced electric field by means of a horizontal electric dipoles 2-5 km in length located both on land and in sea in the coastal zone of the Strait. The measurements confirmed the theoretically predicted by means of numerical modeling amplitude of motionally induced signal. The potential difference measured at the coast of the Strait of the White Sea is dominated by tidally driven signal with the amplitude about 4-8 mV/km on land and about two times greater in the sea. The electric response from quasi-stationary coastal current could be expected to be of about 0.5-2 mV/km. The results of the numerical modeling and the experimental data obtained on south-eastern shore of the Strait of the White Sea support the possibility of the application of on-land measurements of low-frequency electric field in the coastal zone for studying the spatial and temporal variability of integral parameters of sea currents. 2.8p FACTORS EFFECTING ON MOTIONALLY INDUCED ELECTROMAGNETIC FIELD IN GLOBAL SCALE I.V. Yegorov and N.A. Palshin Shirshov Institute of Oceanology, Moscow, Russia yegorov@geo.sio.rssi.ru Growing importance of long-term EM field measurements in oceanographic studies brought forth the question of what are the main factors effecting on spatial and temporal distribution of motionally induced electromagnetic field in global scale. Numerical modeling is essential and effective tool for such study. Multi thin- sheet approach based on finite element method was used for simulating electromagnetic field. Several models are needed for numerical simulation. Near-surface layer conductance model which includes sea-water, bottom sediments and rough estimation of continental sediments and annual/seasonal mean depth-averaged water velocity models based on the global model developed in Shirshov Institute of Oceanology by K. Lebedev were constructed. All the models used have one degree resolution. The effect of following factors were studied: (1) near-surface layer heterogeneity with and without bottom and continental sediments, (2) integral resistivity of the lithospheric mantle, (3) seasonal variability of the global circulation of the World Ocean, including seasonal variability of water conductivity in shallow basin, (4) ionospheric currents inducing significant long-period electromagnetic fields. The most important factor is the combined effect of conductance of near- surface layer and integral resistivity of the lithospheric mantle. Additional studies needed to estimate the influence of lithospheric mantle heterogeneity. Seasonal variability of the water mass circulation could also produce significant response in electromagnetic field, especially within areas where western boundary currents exists. 2.9p ON THE DEPENDENCE OF LARGE-SCALE OCEAN GENERATED ELECTRIC CURRENTS AND MAGNETIC FIELDS ON THE NON-DIPOLE COMPONENT OF THE EARTH'S MAIN MAGNETIC FIELD Robert H. Tyler Applied Physics Laboratory, University of Washington, 1013 NE 40th St., Seattle Washington, 98105-6698, USA tyler@apl.washington.edu We present a dynamo theorem of significance to magnetic fields generated by ocean flow. The theorem states that geostrophic flow in a {thin-shell} ocean and in the presence of an axis aligned main dipole magnetic field will generate only magnetic fields which are entirely confined to the ocean. As a corollary, the magnetic fields generated by geostrophic flow which reach outside of the ocean are dependent on the non-dipole component of the main field. 2.10p MONITORING OF MOTIONALLY INDUCED VOLTAGES IN THE BALTIC SEA Peter Sigray Stockholm University, Meteorological Institution, Svante Arrheniusväg 21a S-106 91 Stockholm, Sweden peters@sto.foa.se peters@misu.su.se The integrated flow at two regions in the Baltic Sea has been measured by utilising the method of motionally induced voltages. The measurement of the cross-channel voltage in the Oresund, between Denmark and Sweden, showed that a barotropic situation can be studied by direct observation of the induced voltage. During the trials the north-going stream out from the Baltic Sea was weakening. Both the tidal component as well as the integrated flow were clearly observed. A continuos monitoring of the salinity and temperature revealed that a salt wedge was present in the region. Comparison with ADCP-data showed that in spite of this wedge the method is robust and gives an accurate estimate of the flow. It was furthermore observed that the tide propagates 'out' from the Baltic Sea towards Kattegatt. By combining the signals of a near-surface mounted rotor-type current meter and the induced signal the salt wedge structure was determined. A second separate measurement between Gotland and the mainland was performed by a parasitic use of an in-use fibre optical cable. The mantle of the cable was grounded on the island. The voltage difference has been recorded during one year. The correlation between the induced electrical field and wind data indicates that the flow is mainly wind driven, as expected in shallow water. 2.11p INDUCTION STUDY USING SQ HARMONIC DERIVED FROM JASC VOLTAGE D.Yu. Abramova(1), L.L. Vanyan(2), N.A. Palshin(2), E.P. Kharin(3), and H.Utada(4) (1) Institute of Terrestrial Magnetism, Ionosphere and Radio Wave Propagation, Troitsk, 142190, Russia (2) Shirshov Institute of Oceanology, Moscow, 117218, Russia (3) World Data Center for Solar-terrestrial Physics, Moscow, 117296, Russia (4) University of Tokyo, Japan u10599@dialup.podolsk.ru The submarine telecommunication cable JASC was replaced by the modern one and was transferred for the scientific research in 1996. Till now the comprehensive dataset of the natural electric potential measurements was created and a good opportunity for MT study the region of the Sea of Japan was appeared. This paper reports first results of an induction study using the solar daily variations which prove to be 90% of measured field energy. For spectra analysis the cable measurements during nine months of 1997, from February to October, have been used looking at the information about Five International Quiet days of each month. Five main daily harmonics have been picked out from the experimental data rather reliably because the short period variations had filtered out by the sea water layer. Geomagnetic data are obtained from Kakioka observatory for the same time. Because of the normal telluric field from the JASC cable is significantly distorted by the coast effect we can use TM-mode field which sensitivity to the resistive lithosphere is high. So a first order approximation MT response was calculated for the Sea of Japan. We compare our preliminary results with previous ones based on continuum spectra produced by the essentially different sources of field. 2.12p ELECTRICAL CONNECTION OF THE OCEAN AND MANTLE AT A SUBDUCTION ZONE: MODEL STUDY FOR THE ANDES AT 21S R.L. Evans(1), A.D. Chave(1), and J.R. Booker(2) (1) Woods Hole Oceanographic Institution, Woods Hole, MA 02543, USA (2) University of Washington, Seattle, WA 98195, USA revans@whoi.edu There is general agreement that water carried in the downgoing hydrated sediments and oceanic crust plays an important role in the volcanic processes beneath convergent margins. However, the amount of water carried into the system, the lateral pathways which it takes, and the depths over which it is released into the mantle wedge are largely unknown. Water has a substantial impact on mantle conductivity at very low concentrations, and is widely invoked to explain the 1-3 order of magnitude increase of measured oceanic mantle conductivity over that predicted for a dry olivine mantle. This suggests that details of the connectedness of the ocean to the mantle through the subducted slab may have a profound effect on the MT response both on- or off-shore. A 2D model study was carried out for simplified end member models of the marginal structure at 21S. The background model was based on results from the ongoing German effort in Chile and Bolivia. The key parameters to be varied are 1-the resistivity of the sediment and crust on the Nazca plate offshore, 2-the resistivity of the subducted Nazca crustal layer to about 40 km depth, 3-the extension of this unit to 80 km, 4-the resistivity of the mantle wedge at 60-80 km depth, and 5-the connectedness of these units to the HCZ extending from the Precordillera to the Altiplano. The first of these is not important. Changing the second parameter from a background 1000 ohm-m to 50 ohm-m, which yields a conductance comparable to that observed in EMSLAB, has a dramatic effect on the offshore TE and TM mode response, with the phase changing by as much as 50 degrees or 20 times the standard error with which it can be measured. The effect onshore is much more subdued and barely detectable. Changing parameters 3-5 has a comparable influence on the offshore response with systematic changes in the shape and size of the TE and TM responses depending on the size and location of the electrical connections in the mantle. However, the offshore data are only weakly dependent on the connections at points east of the Precordillera, and hence will not be strongly affected by alongstrike variations in this region. This model study shows that offshore MT data are very sensitive to electrical connections between the ocean and mantle, while land data cannot sense these effects. These connections are diagnostic of the pathways by which fluid reaches the sub-Andean mantle and their conductance is diagnostic of the amount of water present, and hence of its flux. The effect is dramatically larger at 21S compared to EMSLAB because of the extensive sediment cover at the latter site and the near lack of sediment offshore Chile. 2.13p ALONG RIDGE CHANGES IN THE MANTLE ELECTRICAL CONDUCTIVITY STRUCTURE BENEATH THE EPR BETWEEN 17 AND 15 SOUTH P. Tarits(1), A.D. Chave(2), and R. Evans(2) (1) UBO/IUEM, Place Nicolas Copernic, F-29280 Plouzane, France (2) WHOI, Woods-Hole, MA, USA tarits@univ-brest.fr The EM component of the MELT experiment was aimed at measuring the electrical conductivity structure beneath the southern EPR to depths of several hundreds kilometers below the seafloor. Electrical conductivity is expected to be strongly influenced by the presence of interconnected melt, even at low melt fractions, and is also an indicator of other key mantle properties such as its thermal state and water content. Magnetotelluric (MT) stations were deployed along two E-W lines. The southern line crosses the EPR at 17S on a magma-rich segment of ridge crest, with the MT sites extending 200 km either side. The second line crosses the EPR at about 15-45S to the north of an OSC on a ridge segment which is magma starved with the MT sites extending 100 km either side of the axis. The bi-dimensional (2D) analysis of the magnetotelluric (MT) data acquired along the southern line reveals an asymmetric resistivity structure, with lower resistivity to the west of the ridge. The uppermost 100 km of mantle immediately to the east of the ridge is consistent with a dry olivine resistivity structure indicating a mantle depleted of melt and volatiles. Mantle resistivities to the west of the ridge are consistent with a low melt fraction (about 1-2% melt) distributed over a broad region and extending to depths of about 150km. This difference may be the result of asymmetric spreading rates and a westward migration of the ridge axis, and suggests distinct styles of melt formation and delivery in the mantle beneath the two plates. The 2D mantle conductivity model resulting from the analysis of the MT data along the northern line shows differences with the southern line model. In the West the mantle seems more conductive in the North, while in the East the mantle seems more conductive in the South. 3D modelling is used to examine the sensitivity of the data to the 3D structure suggested by the comparison between the southern and northern models. Implications for the mantle will be considered and the results compared to the seismic and geoid data. 2.14p IMAGE OF THE MANTLE BENEATH THE ACTIVES HOTSPOTS Rita Nolasco(1) and Pascal Tarits(2) (1) Centro de Geofisica da Universidade de Lisboa, R. Escola Politecnica, 58, 1269-102 Lisboa, Portugal (2) Universit de Bretagne Occidentale, UMR, Domaines Oceaniques, Place Nicolas Copernic, F-29280 Plouzane, France dubert@teleweb.pt Seafloor magnetotelluric instruments data were used to image the mantle beneath the active Teahitia hotspot, Tahiti. The inversion of the data shows a slightly higher conductivity, relative to a reference site located away from the hotspot, down to 130km depth beneath the active area Southeast of Tahiti, underlain by a more resistive structure. There is also a suggestion for a change in conductivity in the 400-450km depth range, which is consistent with elevated temperatures. This result is consistent with a mantle plume of limited extent, less than 150km, located near the edge of the Teahitia hotspot. The magnetotelluric data provide no evidence for lithospheric thinning or for a strong thermal influence over a large area, but the presence of fluids over a large area seems to be needed to explain the resistivity profile. The presence of melt at the lithosphere and melted and volatile depleted fraction of the mantle between 130 and about 300km depth are a possible explanation for the data over the active area Southeast of Tahiti. The analysis of magnetotelluric and deep geomagnetic sounding data from another hotspot area, Hawaii, show also a lithosphere with melt and a melted and volatile depleted fraction of the mantle could be a possible explanation for these data. 2.15p SOURCE EFFECTS ON GEOMAGNETIC DATA FROM TAHITI AREA Rita Nolasco(1) and Pascal Tarits(2) (1) Centro de Geofisica da Universidade de Lisboa, R. Escola Politecnica, 58, 1269-102 Lisboa, Portugal (2) Universit de Bretagne Occidentale, UMR, Domaines Oceaniques, Place Nicolas Copernic, F-29280 Plouzane, France dubert@teleweb.pt The seafloor magnetic data in the Tahiti hotspot area have been analysed and the transfer functions between the vertical and horizontal magnetic components, TFV, were computed using a robust remote reference method. We found that the TFV response is sensitive to the choice of a remote reference site. Furthermore, there is a substantial difference between the results obtained with daytime and nighttime data. This suggests significant source field effects on the TFV response, at low and medium periods. A possible cause could be related to the proximity to the equatorial electrojet. However, the modelling of the equatorial electroje show that is effect is not adequate to explain the observed features in the TFV responses. In addition, at long periods, more than a few hours, TFV have a strong polarisation, which is not compatible with the magnetotelluric data. We show that this polarisation is not induced by local 3D effect. This suggest the presence of source effects, oceanic or exospheric?, different at short and long periods. 2.16p INVESTIGATION OF CRUSTAL FLUIDS AT THE LUCKY STRIKE HYDROTHERMAL SITE USING CONTROLLED SOURCE ELECTROMAGNETIC SOUNDING L.M. MacGregor(1), M.C. Sinha(2), F. Santos(3), M. Miranda(3), A. Soares(3), J. Luis(4), N. Lourenco(4), and A.H. Flosadottir(5) (1) Bullard Labs, Univ. of Cambridge, UK (2) School of Ocean and Earth Sciences, Southampton Oceanography Centre, UK (3) Univ. of Lisbon, Portugal (4) Univ. of Algarve, Portugal (5) HALO Ltd. Reykjavik, Iceland mcgregor@esc.cam.ac.uk agusta@pmel.noaa.gov In September 1999 a marine controlled source electromagnetic (CSEM) sounding experiment was performed on the Lucky Strike segment of the Mid-Atlantic Ridge, where both diffuse and high temperature venting have been observed. The Lucky Strike segment has been extensively investigated by sonar, diving and sampling studies, which have yielded excellent data on the geological and geochemical characteristics of the hydrothermal system. However there have been few studies targeted at the large scale properties of the crust through which the fluids flow. Marine CSEM sounding uses artificially generated electromagnetic signals transmitted from a source to an array of seafloor receivers to study resistivity structure on a crustal scale. The CSEM experiment is designed to determine the distribution of electrical resistivity in the crust beneath the Lucky Strike segment, and hence to investigate the porosity and permeability structure of the hydrothermal system, the temperature and depth of penetration of hydrothermal fluids, and the presence or absence of a magmatic heat source beneath the segment. Although the CSEM data alone cannot unambiguously resolve these questions, when combined with existing geophysical and geochemical data the constraints provided by the resistivity structure will provide new insights into the in situ physical properties and plumbing of an active hydrothermal system.