10.1p THE SEA BOTTOM AND LAND ELECTROMAGNETIC STUDIES IN BALCAN- AEGEAN REGION. Authors: Abramova L., Abramov Yu. and Shneyer V. Institute of Geoelectromagnetic Investigations of United Inst. of Physics of the Earth, 142092 TROITSK, Moscow region, box 30, RUSSIA e-mail: abramova@geo.igemi.troitsk.ru This paper presents the results of different marine magnetovariational observations in a vast region, including the coast zone of the Crimea and Caucasus, the Bulgarian continental shelf and water area of the Black and Aegean seas. EM measurements have been made at about 80 points in Bulgaria, Crimea and at the adjacent Black Sea shelf and at the bottom of the deep part of Aegean sea. Variations in the magnetic field were measured with the 3-component bottom flux- gate and buoyant proton variometers (at two horizons). As a result, electrical conductivity functions in the period range of 10 min to3-4 h (and occasionally up to 24 h) were calculated. Non uniform thin-sheet interpretation and 2D models are applied to invert this data. Following results about conductivity distribution were obtained: i. There exists the crust conductivity layer which thickness is about of 5 km, resistivity 4-5 Ohm m, conductance is approximately 1000 S. Its presence connected as assumed with thick series of sediment rocks buried here and points to the allochthonous origin of Rhodope. To some Bulgarian geologists views autochthonous in this shear is the thick layer of karstified marbles what is regarded as a large deep bedding artesian basin with hydrothermal karst. This layer is confirmed also by seismic data the most clear in the Rhodope zone and in the Balkan Alpine zone. ii. The asthenosphere layer, which begins from 50 km depth and have the effective conductance Sa=3000 S was discovered in upper mantle. It degrades both towards Moesian plate to the North and in the Aegean sea to the South before the Cretan depression. Both the electromagnetic measurement data themselves and their analysis determine the asthenosphere conducting zone existence under the Aegean sea, South part of the Bulgarian shelf of the Black sea and Rhodope massif. This result conforms to the high values of the heat flow observed here and is in concordance with modern knowledge about the deep structure of the neighbouring regions. 10.2p CALCULATING THE ELECTRIC AND MAGNETIC FIELDS AT THE SEAFLOOR DUE TO A TWO-DIMENSIONAL IONOSPHERIC CURRENT Authors: David Boteler (1), Risto Pirjola (2) and Ari Viljanen (2) (1)Geological Survey of Canada, Geomagnetic Laboratory (1) Observatory Crescent, Ottawa, Ontario K1A 0Y3, Canada phone: +1-613-8372035 fax: +1-613-8249803 email: Boteler@geolab.emr.ca (2)Finnish Meteorological Institute, Geophysical Research Division, P. O. Box 503, FIN-00101 Helsinki, Finland phone: +358-9-19294652, +358-9-19294668 fax: +358-9-19294603 e-mail: Risto.Pirjola@fmi.fi, Ari.Viljanen@fmi.fi To be able to estimate geomagnetic effects on submarine cables, it is necessary to know the electric field produced at the seafloor during geomagnetic disturbances. In this paper we derive relations between the seafloor electric field and surface magnetic field produced by a source field with E-polarization. This relationship can be expressed as an inverse Fourier integral over the wave number or as a spatial convolution and depends on the depth and conductivity of the seawater and on the properties of the basement. In this study the source is a two-dimensional ionospheric current, and the layered earth below the seawater is described by a surface impedance. Input data are provided by the magnetic recordings collected by the IMAGE magnetometer array in northern Europe. They are used to calculate the electric fields that would be produced on the continental shelf (depth of 100 m) and at a deep ocean seafloor (depth of 5 km). It is seen that the dependence of the electric field on the period of oscillation is weaker for the deep ocean. Also, at long periods (10000 s) the results for the two-dimensional source deviate from the results obtained using the plane wave assumption. 10.3p ELECTROMAGNETIC ASSESSMENT OF OFFSHORE METHANE HYDRATE DEPOSITS: FIRST SURVEY RESULTS Authors: G. Cairns (1), J. Yuan (2), R.N. Edwards (2), H. Utada (3), H. Toh (4) (1) Universite de Bretagne Occidentale, IUEM, UMR "Domaines Oceaniques" Place Nicolas Copernic, F-29280 Plouzane, France phone: 33 2 98 49 87 09 fax: 33 2 98 49 87 60 e-mail: cairns@sdt.univ-brest.fr (2) University of Toronto, Department of Physics 60 St. George St., Toronto, Ontario, Canada. M5S 1A7 (3) University of Tokyo, Earthquake Research Institute Yayoi 1-1-1, Bunkyo-ku, Tokyo 113 Japan (4) University of Tokyo, Ocean Research Institute, Japan Methane hydrates are slushy crystalline mixtures of water and methane which occur worldwide in sediments beneath the seafloor. It is estimated that the total reserves of seafloor methane hydrates exceed 10^16 Kg. Since each volume of hydrate can yield up to 164 volumes of gas, these reserves represent a very important natural energy resource. While the base of hydrate deposits stands out on seismic sections as a bright reflection, the diffuse upper boundary is not well delineated. Estimation of the total mass of hydrates is therefore not easily accomplished by seismic surveys alone, and generally requires complimentary information. As methane hydrate is electrically insulating, its presence in porous sediments causes an increase in the formation resistivity. Electrical data can therefore aid in the evaluation of this resource. Using a recently developed seafloor transient electric dipole-dipole system, an experimental survey was conducted in the summer of 1997 off the coast of Vancouver Island. Electromagnetic soundings were made between a towed transmitter dipole and autonomous seafloor receivers at ranges of 1 km to several hundred meters. We present the initial results of this survey and our assessment of the methodology used. 10.4p VISUALISATION OF 3D MAGNETOTELLURIC DATA USING APPARENT RESISTIVITY AND APPARENT PERMITTIVITY TENSORS Authors: T. G. Caldwell(1), H.M. Bibby(2) and W.I. Reilly(3) (1) Institute of Geological and Nuclear Sciences, Wellington, New Zealand. e-mail: g.caldwell@gns.cri.nz (2) Institute of Geological and Nuclear Sciences, Wellington, New Zealand. e-mail: h.bibby@gns.cri.nz (3) Department of Surveying, University of Otago, Dunedin, New Zealand. e-mail: ireilly@southern.co.nz MT data are usually analysed by identifying the Cartesian co-ordinate system that minimises the trace of the impedance tensor and results in the closest approximation to a 2D situation. Where the situation is 3D, this approach neglects some of the information available. Any resulting interpretation will also depend on the co-ordinate system. Using an analogy between the EM field measured at the surface and the behaviour of an EM wave in an anisotropic material that is both polarisable and conductive an alternative method of analysis is possible. Two real tensors, representing the apparent resistivity and apparent permittivity are defined. The apparent permittivity represents the gross capacitative behaviour of the earth near the measurement site just as the apparent resistivity represents the gross conductive behaviour. This analysis makes explicit that, in general, two independent sets of principal axes characterise MT data and resolves the difficulties that arise in attempting to define a single preferred co-ordinate system in a 3D situation. Seven co-ordinate invariant parameters can be derived from the tensors, each invariant corresponding to a feature in the system of two ellipses that can be used to represent the tensors graphically. The ellipses provide a compact way of visualising MT data. 10.5p TEM MEASUREMENTS IN BOLIVIA COMPARISON OF DIFFERENT FAST INTERPRETATION METHODS Authors: S. Dautel, S. Greinwald, Ch. Grissemann, A. Hoerdt, H. Thern,L. Winkelmann A geophysical survey using Audiomagnetotellurics (AMT), Transient-Electromagnetics (TEM) and Induced-Polarisation (IP) measurements was carried out near the former silver mine at Pulacayo, Bolivia, in 1995 to support a structural geology investigation of the area. Different techniques for data interpretation were compared. For the TEM soundings uniform transmitter loops (size of 800 m*400 m) were placed along 4 profiles (total length 15.3 km) with a separation of 700m. For each transmitter location measurements were taken at 15 receiver positions with a spacing of 50 m along the profile. The data was interpreted with two different 1D-Imaging methods, 1D-Marquardt-Inversion and 2D Imaging. Although good results had been obtained with synthetic data, the 1D imaging algorithms turned out to be sensitive to noise. Thus the results did not provide additional information to the apparent resistivity pseudosections. 1D Inversion was the best method to give fast and detailed information of the conductivity distribution of the subsurface. The 2D Imaging showed a rough estimate of the structure without additional information to the 1D Inversion results. The TEM Measurements with 1D Inversion of the data could be reconciled with the geological information and provided new information about the structure of the subsurface to an exploration depth of 1 km. 10.6p MAGNETIC AND ELECTROMAGNETIC INDUCTION EFFECTS IN THE ANNUAL MEANS OF GEOMAGNETIC ELEMENTS Authors: Crisan Demetrescu, Venera Dobrica, Nadejda Ciocoiu Institute of Geodynamics, 19-21 J.L. Calderon Str., Bucharest, Romania e-mail:crisan@geodin.ro It is a well established fact that in the annual means of the geomagnetic elements recorded at observatories as well as in the annual values recorded at repeat stations a solar-cycle-related (SC) variation is present. The SC variation in the annual means of the horizontal and vertical components of the geomagnetic field at European observatories is used to infer information on the magnetic and electric properties of the interior, characteristic to the observatory location, by identifying and analysing the magnetic induction component and respectively the electromagnetic induction component of the SC variation. The obtained results and the method can also be used to better constrain the anomaly bias in main field modelling, and to improve the reliability of secular variation models beyond the time interval covered by data. 10.7p MULTIVARIATE ANALYSIS OF ELECTROMAGNETIC ARRAY DATA SEPARATION OF COHERENT NOISE FROM MT SIGNAL ? Authors: Gary D. Egbert, Markus Eisel, COAS, Oregon State University, Corvallis, OR, 97331, e-mail: egbert@oce,orst.edu Eigenvector decomposition of the array spectral density matrix (SDM) allows a simple test for the presence of coherent noise: if there are more than two large eigenvalues of the (incoherent noise variance normalized) SDM there is coherent noise. Separating the coherent noise from the desired MT signal is often not so simple. We discuss here some approaches to this problem. These include: (1) Use of methods based on canonical coherence analysis to identify subsets of channels over which noise is coherent (or not), (2) assuming smoothness (in frequency) of plane wave source transfer functions to separate MT components from narrow band coherent noise components, and (3) use of the constraint that coherent noise should on average be statistically independent of natural source plane wave variations. We present example applications of these methods to coherent noise contaminated MT datasets from California and Bavaria, illustrating how these methods can be used to clarify the nature of coherent noise, to improve estimates of MT transfer functions, and to provide more realistic estimates of the magnitude of systematic bias errors in MT TFs. 10.8p A STUDY ON TIME STABILITY OF MT AND MV TRANSFER FUCTIONS, VARIANCES AND NOISE Authors: Markus Eisel(1), Gary D. Egbert(1), O. Sierra Boyd(2), H. Frank Morrison(2) (1) Oregon State University, College of Oceanic and Atmospheric Sciences, Ocean. Admin. Bldg. 104, Corvallis, Oregon 97331, e-mail: egbert@oce.orst.edu eisel@oce.orst.edu (2)Earth Sciences Division 90-1116, Lawrence Berkeley Laboratory, 1 Cyclotron Road, Berkeley, CA, 94720, e-mail: sierra@eg426.berkeley.edu hfmengeo@socrates.berkeley.edu In an experiment to study possible correlations the proposed relation between earthquakes and changes in the Earths electro-magnetic field, five-component MT data have been collected continuously since 1996 at two sites in central California. The sites are located along the San Andreas Fault, about 150 km apart with data recorded at both sites simultaneously 40 Hz. This paper investigates the long term stability of MT and interstation magnetic transfer functions, their variances and related parameters. The multiple station approach (Egbert, 1997) is used for the estimation of transfer functions. Using different time windows for averaging reveals the influence of cultural noise and natural signal power peaks and gives suggestions for 'best recording times'. Daily averages from two years of data show that there are no significant seasonal variations of MT transfer functions, though the errors show some variation with season. Interstation magnetic transfer functions reveal - in addition to a gradient in the natural source field - the influence of coherent noise between the two stations, suggesting the existence of a distant, large scale source of cultural noise. Detailed studies of the noise in terms of daily variations and source direction suggest electric trains (BART) as a possible source. 10.9p INDUCTION SOURCES IN THE OCEAN: THE ISO-3D PROJECT Authors: A.H. Flosadottir, D. Greer, A. Junge, J.F. Luis, L.M. MacGregor, J.M. Miranda, S. Riches, F.A.M. Santos, M.C. Sinha, A. Soares, R.H. Tyler Corresponding author: Dr. Agusta H. Flosadottir. University of Washington and HALO, Inc. HALO, Laboratory of Oceanic and Atmospheric Sciences, Austurstraeti 16, P.O. Box 638, IS-121 Reykjavik, Iceland Other contact info: phone: +354-1-525-4951, fax: +354-1-525-4079 e-mail: agusta@pmel.noaa.gov A partnership of researchers in the United Kingdom, Iceland, Portugal, and Germany is in the early stages of a collaborative project aimed at improving the exploitation of observations of electric and magnetic fields in the oceans induced by both artificial and natural sources, and focusing in particular on the extension of existing numerical and experimental methodologies to take account of fully three-dimensional geometries. Among the objectives of this research and technology project, funded by the European Union's MAST-III and entitled ``Applications of 3-Dimensional Electromagnetic by Sources in the Ocean" (abbreviated ISO-3D) are: 1) Development of a numerical computer model and its use in model studies of applications to both physical oceanography and geophysics, including the effects of induction by natural and artificial sources within a three-dimensional ocean and earth, 2) A three-dimensional controlled-source (CSEM) experiment on the Lucky Strike segment of the Mid-Atlantic Ridge south of the Azores, and 3) Modeling and analysis of ocean-tidally generated fields seen in land- and cable-based observations. The poster will provide an introduction to the various aspects of the project and will present preliminary results. 10.10p GLOBAL LIGHTNING TRIANGULATION Author: Dr. Martin Fullerkrug Time variations of the natural electromagnetic environment in the frequency range 4-100 Hz are characterized by the Earth-ionosphere cavity (or Schumann) resonances. These resonances result from the interference of propagating waves within the spherical shell bounded by the conducting Earth and the ionosphere and they are excited by particularly strong lightning flashes within continental thunderstorms mainly ocurring in the tropical belt around the world. Simultaneous recordings of continously sampled time series at several stations around the world are used to triangulate individual lightning flashes. The resulting picture of thunderstorm activity is validated by use of local lightning detection networks and optical satellite observations. The installation of an operational network has scientific applications in the estimation of continental temperature variations, assessing the global NOx production by lightning and sprite location determination. 10.11p APPARENT RESISTIVITY AND PHASE ESTIMATION FROM MAGNETIC FIELD MEASUREMENTS Authors: Mehran Gharibi and Laust B. Pedersen University of Uppsala, Department of Geophysics, Section of Solid Earth Physics, Villavagen 16, S-752 36 Uppsala, Sweden phone: +46-(0)18-471 33 22 fax: +46-(0)18-50 11 10 e-mail: mg@geofys.uu.se The VLF technique is one of the most common methods in electromagnetic surveys for mapping of near surface geological structures on a large scale. The ratio of the vertical magnetic component divided by the horizontal magnetic component (tipper) has been the main indicator of lateral variations of electrical conductivity in qualitative interpretation of VLF data. In order to allow and facilitate a quantitative interpretation of VLF data a very fast iterative method has been developed to transform the real and imaginary parts of the tipper values along a profile over a 2D geological structure into impedances or apparent resistivities and phases. A relationship between horizontal electric field and vertical magnetic field was derived from Maxwell's equation in E-polarization. A full description of disturbed magnetic field components was derived by making use of the fact that the secondary horizontal and vertical magnetic fields are of internal origin, hence form a Hilbert transform pair. Synthetic and field survey VLF data have been used to assess the performance of the method. The results show a remarkable recovery of apparent resistivities and phases in either case as long as the profile is sufficiently long. 10.12p INTERACTIONS BETWEEN MAGNETIC MAPPING AND ELECTROMAGNETIC INDUCTION IN THE EARTH Authors: Adrian P. Hitchman (1), F.E.M. (Ted) Lilley (1) and Peter R. Milligan (2) (1) Australian National University, Research School of Earth Sciences, Canberra ACT 0200, Australia e-mail: adrian@rses.anu.edu.au (2) Australian Geological Survey Organisation Time-varying magnetic fields induced in Earth become part of the total magnetic field measured by aeromagnetic mapping. The CICADA project is concerned with such induction effects, looking both for ways to improve the resolution of the mapping, and to exploit aeromagnetic data for information on conductivity structure. There are presently four main strands to the CICADA project: *Analysis of aeromagnetic crossover misfits for conductivity information. Using a Fourier technique, magnetic daily variations have been reconstructed which show evidence of induced fields. *Global Sq total-field curves. Because the diurnal variation in the total-field is of prime importance in aeromagnetic mapping, type curves for Sq, for the whole globe, have been derived from the analysis of Campbell for the International Year of the Quiet Sun. *Total-field magnetic amphidromes. Near a conductivity contrast, for particular inclinations of the main magnetic field, total-field fluctuations of period less than one hour may be undetectable with a scalar magnetometer. This phenomenon has been termed a "magnetic amphidrome". *Micropulsations and the coast effect. Because of the importance of micropulsations as a noise source, and the presence of the coast-effect when aeromagnetic surveys take place over continental shelves, three-component observations of the coast effect have been made with a line of magnetometers at right angles to an Australian coast, including a seafloor magnetometer on the continental shelf. (CICADA: Clarifying Induction Contributions to Aeromagnetic DAta) 10.13p MONITORING AN UNDERGROUND GAS STORAGE RESERVOIR WITH TRANSIENT ELECTROMAGNETICS Author: Andreas Hoerdt University of Cologne, Institute for Geophysics and Meteorology, Albertus Magnus Platz, 50933 Koeln, Germany phone: 49-221-470-5477 fax: 49-221-470-5481 e-mail: hoerdt@ageo.uni-koeln.de Natural gas is an increasingly important energy resource in Europe. Imported gas is stored in underground reservoirs, where monitoring of gas volume changes is essential for a proper management. Electromagnetics could help to reduce the number of expensive monitoring wells by monitoring resistivity changes from the surface. Two transient electromagnetic surveys were carried out over an underground gas storage site in France. The aim was to conclude from differences in the data to movements of the gas bubble. Preliminary 3-D modelling, using all available a priori information, indicated that the biggest expected difference in the data occurs in the electric field step responses and is about 5 %. The data acquisition was done with the TEAMEX multichannel acquisition system, with one receiver station every 125 m along a 4 km receiver profile. Parallel and perdendicular electric dipole transmitters were installed every 250 m. The analysis of the differences shows that many transients have a reproducability with errors less than 1 %. However, no smooth, correlated signal which could reveal a change in the gas volume was detected. The most important reasons for this are known systematic errors, and a change in the gas volume which was smaller than anticipated. 10.14p FORTY YEARS OF MTS STUDIES IN UKRAINA Authors: Ingerov A.I., Rokityansky I.I. Geophys.res.lab. "Slavuta", Institute of Geophysics, POB-338/7, Kiev-146, Ukraina. e-mail: earth@igph.kiev.ua In 1958 Dr. Nina V.Lipskaya added two horisontal magnetic components to Earth currents measurements drived since 1957 within the framework of International Geophysical Year in observatories Koretz in Ukrainian shield and Alushta in Crimea. Investigation of the observed MT field behavior led to a set of developments: analysis of MT-anisotropy (1961), coastal effect (1963), correcting shift of statically distorted MTS curves to global data (1971), analysis of geomagnetic variation's anomalies (1969-1975). Carpathian, Kirovograd and Donbass regional electrical conductivity anomalies vere discovered and studied by MVP method. MTS in 60-70-ies were mostly used in sedimentary basins for oil prospecting. The Romanian Vrancea earthquake of 1977 felt in Kiev and Moscow stimulated the great Ukrainian MTS project. In 1980-ies thousands digital MTS were made on all the territory with 40x40km network, along several transukrainian geotraverses (10 km span) and many more detailed observations on foults, deposits, near nuclear plants and so on. The data are processed, a small part is analysed and published. Our intention to make an all-Ukrainian map of crustal conductivity met methodical difficulties. 10.15p HIGH-RESOLVING ELECTROMAGNETIC PROSPECTING IN FRECVENCY DOMAIN Authors: Ingerov A.I.(1),Kusch O.A.(3),Soldatenko V.P.(2), Lozovoj A.L.(2), Vlasov Y.T.(1) (1) GRL Slavuta Co.,Ltd,33,K.Marx Ave, Dnipropetrovsk,320044,Ukraine, Fax (380562) 476777 (2) Institute association "Ukruglegeologia",Ukraine,17 Illicha Ave.,Donetsk, Ukraine (3)National Mining Academy of Ukraine, 19 K.Marx Ave.,Dniepropetrovsk,320027, Ukraine e-mail:root@slavuta.dnepropetrovsk.ua By application of electrical prospecting with a controlled source the use of enterference extremes of amplitude and phase frecvency caracteristics is a reserve for raising the exactitude of determination of the depth of contrast electrical boundaries in geological section.These extremes are caused by interference of direct EM-wave propagating in air and E wave (S) reflected (refracted) on some rigid boundaries in the earth. The coordinates of interference extremes amplitude and phase curves FDEM are closely connected with the depth of the boundaries in geological section. The practice of field works by the method FDEM and the interpretation with the use of coordinates of interference parameters has found application in Ukraine by solving main problems of electrical prospecting. Varying with the net of observations, type of FDEM set and changeability of boundaries the exactitude of mapping of boundaries is 1-5% of the depth value. There are used, as a rule, 2-3 spacing of sounding or more. To make the results more exact is possible by using as a training sample the data of seismical prospecting on the reference profils or the data of drilling. The ways to high resolving ability of FDEM are: - to mesure not only amplitude but phase parameters as well; - to make ineaserements on different spacing; - to use in interpretation some multidimensional connections of combinations of coordinates of interference extremes. Using multychanel equipment as Phoenics and EMI for FDEM survey are vary perspective in nearest future. 10.16p COMPARING SIMULTANEOUS AMT OBSERVATIONS FOR SITE SEPARATION BETWEEN 1 AND 20.000 KM Authors: A. Junge (1), K. Whaler (2), M. Fuellekrug (1), G.Dawes (2), U. Weckmann (1) (1)Institut fuer Meterologie und Geophysik, Universitaet Frankfurt am Main, Feldbergstr. 47, D-60323 Frankfurt am Main, Germany e-mail: junge@geophysik.uni-frankfurt.de (2)Department of Geology and Geophysics, The Universtiy of Edinburgh, West Mains Road, Edinburgh EH9 3JW, Scotland UK e-mail:kathy.whaler@ed.ac.uk During the last year we have collected (and are still collecting) a number of simultaneous observations of magnetic and telluric field variations in the period range between 1 msec and 1000 sec at various field sites. The sites were seperated by several kilometers distance (Spessart Mountains and Rhenish Massiv, Central Germany), several hundreds of kilometers (Germany, Scotland), several thousands of kilometers (Germany, California) and several tenthousands of kilometers (Germany, New Zealand). The measurements are primarily used by one of the authors (MF) to study the propagation of electromagnetic waves in the Earth-Ionosphere-Cavity induced by lightening, but they may also be used to study remote reference techniques for different distances. Due to GPS time sychronization of the SPAM Mk III data loggers being mostly used for recording the data, the accuracy of the absolute time signal was at least +- 1 msec. We shall show dependencies of various inter-station cross spectra of distance, daytime and source field orientation and we shall discuss the use and the limits of long distance remote reference techniques. 10.17p MAGNETIC AND MAGNETOTELLURIC SOUNDINGS ON MONTECRISTO ISLAND (ITALY) Author: Arnfried Magunia Institut fur Meteorologie und Geophysik J.W.-Goethe-Universitet Feldbergstr. 47 D-60323 Frankfurt a.M. phone: 0049-69-79824908 fax: 0049-69-79823280 e-mail: magunia@geophysik.uni-frankfurt.de In 1997, a joint research project which involved Italian and German institutes was conducted within the activities of GEOSTAR. Magnetic and (audio-)-magnetotelluric measurements were performed on and around Montecristo Island (Italy) by using ground-based and sea bottom sensors. It was the primary research objective to develop a quantitative model of the electrical conductivity distribution beneath the Tyrrhenian Sea at the westerly extent of the Tuscan geothermal anomaly. 10.18p 3D MODELLING OF THE ELECTROMAGNETIC ANOMALIES OBSERVED IN ZONES WITH MAGMATIC ACTIVITY Authors: Patricia Martinelli and Ana Osella Dto. de Fisica - Facultad de Ciencias Exactas y Naturales - Universidad de Buenos Aires, and also at CONICET Ciudad Universitaria - Pab. 1 - Dto. de Fisica, (1428) Buenos Aires - Argentina e-mail: pmartine@df.uba.ar The electromagnetic anomalies observed in zones with magmatic activity often exhibit 3D effects. The magnitude of these effects varies, depending on the characteristics of each particular structure, but there are certainly cases for which an accurate modeling of the electromagnetic response only can be achieved using 3D methods. Various methods have been developed for 3D magnetotelluric modeling. They are generally based on the Integral Equations technique or on Finite Differences. In addition, the authors have formulated an alternative one, using a Rayleigh-Fourier approach, which is specially adequate to model multilayered structures. In this work, this last method is applied to analyze the characteristics of the magnetoteluric response of different possible conductivity distributions associated with plutonic or volcanic activity. The study contributes to the knowledge of the importance of 3D effects and the conditions under which a conducting zone can be detected. 10.19p MAGNETOTELLURICS IN THE FROZEN NORTH: MEASUREMENTS ON LAKE ICE Authors: Gary McNeice (1), Alan G. Jones (2) (1)Phoenix Geophysics, Scarborough (2)Geological Survey of Canada, Ottawa Making magnetotelluric measurements during the deep Canadian winter is hampered by poor electrode contact because of the frozen ground. To test a novel acquisition configuration, an experiment was conducted on a frozen lake with electrodes deployed on the ground on the shore, in the lake ice, floating below the lake ice, and on the lake bottom. As well, magnetic field measurements were made both on the lake ice and on the shore. Our conclusions from this test survey are: 1. Installation of magnetic sensors on the ice surface of a lake can result in the introduction of significant low frequency noise. This noise is most likely the result of expansion and contraction of the lake ice, movement of water below the ice surface and possibly wind noise. Magnetic sensors must be installed on land to avoid this problem. 2. The procedure of installing telluric electrodes on a lake bottom appears to compare favorably to measurements made on land. The telluric noise floor estimates found though parallel sensor tests support this conclusion. Accordingly, a split-MT system is reqired that measures the magnetic fields on the land, and the electric fields on the lake. We observed large static shift effects on the lake-based measurements compared to the land soundings. This suggests that the lake bottom measurement procedure may be prone to significant telluric distortions. This problem however may be solely related to conditions at the selected test site. Following the test survey, a total of eleven measurements have been made along a portion of the 600-km-long winter road from east of Yellowknife to the Lupin gold mine on Contwoyto Lake. These responses are of excellent quality. Severe static effects are seen for responses from smaller lakes, suggesting that the effect is likely to be lake-size dependent. 10.20p 2-D ELECTROMAGNETIC MODELLING WITH A DIPOLE SOURCE AND TOPOGRAPHY BY FINITE-ELEMENT METHOD Authors: Yuji Mitsuhata(1), Koichi Matsuo(2) and Masato Minegishi(2) (1)Geological Survey of Japan, Japan (2)Japan National Oil Corporation, Japan The controlled-source electromagnetic methods have great possibiliti es. Especially, the controlled-source MT (CSMT) and long-offset transient EM (LOTEM) methods have been applied for deep targets. However, the use of the artificial source causes the distortions due to the anomalies existing near the source or between the source and receiver. In this paper, I employ a pse udo-delta function to describe a dipole source, and demonstrate its efficien cy. Owing to the pseudo-delta function, we can put the dipole source everywh ere, and we do not need to consider the 1-D structure as a primary model at all. In addition, I show the topography effect, which becomes easy to be inc luded because of the complete freedom from the 1-D structure and application of the isoparametric elements. I focus attention on the configurations of th e CSMT and LOTEM surveys with the horizontal electric dipole source. In orde r to check the validity of the code, I compare the results with them obtaine d by other numerical codes. 10.21p A NEW METHOD TO COMPENSATE FOR BIAS IN SINGLE-SITE MAGNETOTELLURICS Author: A. Mueller GeoForschungsZentrum Potsdam, Germany Different methods to compensate for bias in magnetotelluric impedance estimation have been described using additional data from a "noise-free" reference station (Gamble et al., 1979, Larsen et al., 1996). But often there is a lack of reliable reference data. In these cases one might calculate magnetotelluric impedances either by minimisation of noise in (1) the electric or in the (2) magnetic field components. Attempts have been made to get a weighted result using these two different estimates (Sims et al., 1971). Because the new method assumes a special noise model as well as an approximation for the impedance tensor's structure it is not generally applicable. Nevertheless the method has been successfully applied in many cases. It gives a measure for the amount of noise in either the electric or the magnetic data. With this measure the larger components of the impedance tensor can be corrected for the influence of bias. Examples from magnetotelluric measurements at the Merapi volcano/Central Java as well as from the North Anatolian Fault Zone /Turkey will be given to prove the reliability of the new method. 10.22p PRELIMINARY RESULTS FROM A MAGNETOTELLURIC SURVEY ACROSS THE TAUPO VOLCANIC ZONE, NEW ZEALAND Authors: Y. Ogawa(1), S.Takakura(1), N.Matsushima(1), T.Uchida(1), T. Tosha(1), Y. Nishi (1), G. Caldwell(2), H. Bibby(2) and S.Bennie(2). (1) Geological Survey of Japan, Tsukuba, Japan. e-mail: oga@gsj.go.jp (2) Institute of Geological and Nuclear Sciences, New Zealand. e-mail: g.caldwell@gns.cri.nz The Taupo Volcanic Zone (TVZ) in New Zealand's north island is a 6000 km2 area of recent rhyolitic volcanism. Twenty three high temperature geothermal systems occur in this area discharging 4000 MW of heat. A program of remote referenced MT soundings to investigate the resistivity structure of the deeper parts of the crust and upper mantle commenced in March 1998. Wide-band (0.01s - 1,000s) and long period (20s -13,000s) data are being collected along a 150km long profile crossing the zone of high heat flow. Preliminary (one-dimensional) analysis of the long period TE mode data from 3 sites situated at the eastern end of the profile show that for the 2 sites beyond the eastern boundary of TVZ a resistive (200-300 Ohm-m) upper crust 15-20km thick is underlain by more a conductive lower crustal layer (20-70 Ohm-m). The conductance of the layer is similar at both sites. Within the high heat flow area, upper crustal resistivities are lower (80 Ohm-m) and persist to greater depth (60km). There is also no evidence at this site for the high conductivity layer observed to the east. 10.23p STUDY OF AN UNCERTAINTY OF A SOLUTION OF 2-D INVERSE MT PROBLEM Authors: Petrova I. M., Porokhova L.N. Ulyanovskaya, 1, Research Institute for Physics, St. Petersburg State University, St. Petersburg, Russia, 198904 e-mail: Irina.Petrova@paloma.spbu.ru Ludmila.Porokhova.@pobox.spbu.ru We suggest a method for evaluation of an uncertainty of a solution of 2-D inverse bimodal MT problem. The confidence interval I(z, y) and spreading interval J(z, y) are calculated at each point of the YZ-plane simultaneously with solving of the problem, that is finding of the electrical conductivity. For this purpose the 2-D Frechet kernels are used. The functions I(z, y) and J(z, y) depend on a quality of observed data and do not reflect errors linked with approximate methods of forward problem solution. The product I(z, y)J(z, y) characterizes the uncertainty of conductivity in the local point (z, y). The uncertainty of the problem in the whole is considered as a criterion of the uncertainty at the local points. The algorithm was tested on some models and the reliability of obtained solutions was evaluated. Points at which calculated conductivity differ significantly from its "true" value, fall into the region with large uncertainty. The uncertainty in the top part of the Earth would be decreased at the expense of usage of the periods less than 1s. 10.24p 40 YEARS AGO - FIRST GDS-SITE IN SINAIA Author: Eberhard Ritter GeoForschungsZentrum Potsdam, Adolf-Schmidt-Observatory for Geomagnetism, Lindenstr. 7, D-14823 Niemegk, Germany e-mail: ritter@gfz-potsdam.de During the IGY 1957/58 H. WIESE made his well-known pilot studies concerning Geomagnetic Deep Soundings in Middle and South-East Europe. So he had 9 stations in Romania in the spring-time 1958. One of the sites where he registrated the H-, D-, and Z-components of the geomagnetic field, had been in Sinaia. We look back to this work, compare the results with those obtained later on by D. STANICA and his co-workers and discuss the stability of induction arrows, especially the local time dependence. 10.25p EXTENDED PHYSICS RELEVANT TO EM PHENOMENA (phenomenological review) Author: Rokityansky I.I. Inst.of Geophysics, POB-338/7, Kiev-146, Ukraina Earth lights, luminosity of the atmosphere, cloud's lineaments, anomalies of ionosphere and others are the phenomena that are more frequently observed before, during or after earthquakes. The nature of this phenomena is unclear. Local EM fluctuations, local natural strokes (explosions) with sharp reduction of Earth gravitation force, tornado are one more set of nonexplained phenomena. Quasispontaneous variations of physical parameters of different nature which correlate with each other at distant points and with solar activity, sometimes outstripping it. The variation can be also induced by a local sufficiently strong irreversible process. Any parameter, in particular elasticity of torsion pendulum filament or spring, elasticity of quartz cristall, even distance between quantum levels in the atom, i.e. parameters defining the course of clocks, can vary. There are geophysical anomalies of clock's course. This type of the anomalies can be relatively easily measured, which can give rise to new geophysical method development. The possible hypothetical explanations of the phenomena outlined are discussed in the conclusion. 10.26p THEORETICAL ASPECTS OF THROUGH-CASING RESISTIVITY Authors: B. Sh. Singer and K.-M. Strack CRCAMET, Bldg. E5A, Rm. 348 Macquarie University North Ryde, Sydney, NSW 2109 phone: (612)9850-9281 fax.:(612)9850-8366 e-mail: bsinger@laurel.ocs.mq.edu.au We analyze a basic model of the through-casing resistivity measurement. The model consists of an infinite homogeneous metal casing surrounded by a homogeneous cement annulus and embedded into a homogeneous formation. An analysis shows that the distribution of the current injected into the casing becomes axially symmetric at a distance of just a few radii from the injection point. As a consequence, the number and the physical location of the electrodes used to inject electric current become of no importance if the measurement electrodes are not placed too close to the sensing electrodes. A closed form solution is derived for the distribution of the electric potential at a distance that is small or large compared to the characteristic length. The solution shows that a transition from the cylindrical to the spherical pattern of the current flow occurs at a radial distance controlled by the characteristic length. This phenomenon bounds the "current tube" resistance. The analysis results in an explicit expression for the characteristic length and the k-factor of the through casing resistivity measurement. The last parameters depends on the formation resistivity. This dependence should be taken into account in the quantitative interpretation of through-casing resistivity logs. 10.27p MAGNETOTELLURIC TRANSFER FUNCTIONS ROBUST ESTIMATION WITH HIGH BREAKPOINT. Author: M.Yu.Smirnov Institute of Physics, St.Petersburg University, Ulyanovskaya 1, Petrodvorets, St.Petersburg, 198904, Russia. e-mail: smirnov@snoopy.phys.spbu.ru The algorithm of magnetotelluric data processing with use of robust procedures is developed. It is based on Zigel estimation of regression problem, that have a maximal breakpoint equal 50% and minimize maximal bias. It is calculated using algorithm of repeated medians. To increase efficiency for short samples, the Zigel estimation is supplemented by a regression M-estimation and undertakes as initial approximation. The spectral transformation in algorithm is carried out with use of cascade decimation procedure . The decimation factor is chosen equal 3. Further with chosen length of a time window (N = 64) the whole set of the data is treated. For each data segment is carried out: 1. mean removing ; 2. processing by a spectral window; 3. FFT. For the further statistical processing only the spectrums past removing on partial coherence functions for main impedance tensor components are saved. 10.28p FIRST RESULTS OF HIGH-FREQUENCY MAGNETOTELLURIC INVESTIGATION IN POLAND M.Stefaniuk*, T.Czerwinski**, A.Wajda**, T.Mrzyglod** *University of Mining and Metallurgy, al.Mickiewicza 30, 30-059 Cracow, Poland e-mail: stefan@geolog.geol.agh.edu.pl **Geophysical Exploration Company, ul. Jagiellonska 76, 03-301 Warsaw, Poland In 1997, the Geophysical Exploration Company, Warsaw, started magnetotelluric investigation in the Polish Carpathians with the use of the MT-1 system produced by the Electromagnetic Instruments Inc., Richmond, California. Basic measurement and interpretation problems were related with strong artificial EM noise and complex geological structure of the region. To eliminate noise effect on MT sounding curves, magnetic field remote referencing was employed in data measurement and processing. However, as a result of the predominant noise signals standard remote reference data processing not always proved effective. Therefore, results of some computations were verified using the Larsen robust processing and forward magnetotelluric modelling for soundings made close to boreholes. The objective of MT investigation was to identify the topography of the high-resistivity roof and assess resistivity distribution in the basement. A field site array was transverse to the strike of flysch beds and consisted of four pairs of mutually perpendicular electric dipoles arranged in a short line, and a pair of magnetic sensors. Measurements were made along two lines ca 20 km long, with a spacing of about 1.5 km. 1D and 2D inversion algorithms, and 3D forward modelling were used in sounding data interpretation. 10.29p THE TRANSIENT FIELD OF A PERFECTLY CONDUCTING HALF-PLANE IN A CONDUCTING HORST MEDIUM Author: Peter Weidelt Institute of Geophysics and Meteorology Technical University of Braunschweig, Germany D-38106 Braunschweig, Germany e-mail: weidelt@geophys.nat.tu-bs.de An extremely simple exact solution is derived for the quasi-stationary time derivative of the magnetic field scattered by a perfectly conducting halfplane, which is embedded in a uniform conducting host and energized by a unit step impulse of an arbitrarily oriented magnetic dipole. In the late time regime, the response of the host shows the familar t**(-2.5) response, whereas the scatttered field of the halfplane decays only as t**(-2). The main purpose of the results may be the validation of numerical codes. The solution is extended to include displacement currents. Unfortunutely the solution now loses much of its simplicity. 10.30p A GENERAL NONLINEAR ESTIMATOR FOR INVERSION OF RESISTIVITY LOGGING DATA Authors: Zhiyi Zhang (1), Alberto Mezzatesta (1), Sven Treitel (2) (1) Western Atlas Logging Services, 10201 Westheimer, Houston, TX 77042,USA e-mail: zhiyiz@sun180.aws.waii.com (2) TriDekon Inc., 6 E. 5 St., Suite 308, Tulsa, OK 74103-4430, USA e-mail: streitel@ionet.net The goal of the inversion of resistivity logging data is to obtain information about resistivity structures which can lead to accurate and reliable estimation of the formation properties. The challenges in inverting resistivity logging data include resolving models with thin layers in high resistivity contrast environment, handling of data noise, and the incorporation of a priori information about the model. Bayes' inversion has been applied to solve geophysical inverse problems with success, and in this paper we use Bayes' inversion technique to solve the inverse problem in the borehole environment. In applying Bayes' inverse theory to geophysical problems, the question arises which type of probability density function (pdf) is to be used for data noise and a priori information. The three most commonly used pdfs are the Gaussian, the double exponential, and the uniform distribution. Each of them has advantages and finds applications in the inversion of geophysical data. In this paper, we propose a general pdf which includes the Gaussian and double exponential distributions as special cases and develop a Bayes' type of inversion algorithm using this proposed new pdf. The numerical examples show that this new inversion algorithm can take advantage of a priori information and generate results which are geologically and petrophysically meaningful. 10.31p Global Optimisation of Time Domain Electromagnetic Data using Very Fast Simulated Annealing S.P. Sharma1, P. Kaikkonen1 and S.K. Verma2 1 Department of Geophysics, University of Oulu, FIN-90570 Oulu, FINLAND 2 National Geophysical Research Institute, Hyderabad - 500 007 (A.P.), INDIA e-mail: sharma@babel.oulu.fi Pertti.Kaikkonen@oulu.fi postmast@csngri.ren.nic.in ABSTRACT Efficacy of the global optimisation technique is demonstrated in the inversion of time domain electromagnetic data. Transient EM responses observed using a coincident loops system along several profiles and different time channels over a plate like conducting body are inverted using very fast simulated annealing (VFSA) as an optimisation tool. Three time channels (1, 4.5 and 10 ms) for each profile are considered in the inversion. Study reveals that only one run of global inversion considering a single multi-channel profile is not enough to resolve all the 9 model parameters of a plate-like conducting body. However, the global inversion of a single multi-channel profile with several runs yields a mean model such that it is quite close to the true model. Considering many profiles and time channels together in the global inversion can yield reliable estimates of all the parameters. Computationally this is not an efficient procedure. Analysis of the results shows, however that two distant profiles are enough to yield all the model parameters reliably even after one run of the global inversion. Both noise-free and noisy synthetic data are used in inversion. Finally field data are also inverted to study the efficacy of the global inversion. Session 10 : other contributions Presentation: Oral or poster 10.32p Global optimization of transient EM data in conducting surroundings using free-space plate model S.P. Sharma and P. Kaikkonen Department of Geophysics, University of Oulu, FIN-90570 Oulu, FINLAND e-mail: sharma@babel.oulu.fi Pertti.Kaikkonen@oulu.fi ABSTRACT Free-space transient electromagnetic (TEM) formulation for a plate-like conducting body is used to invert measured data affected by the conducting surroundings. Global optimization results reveal that inversion of the data affected by the finite counductivities of the surroundings using a free-space formulation could be unreliable. Good fittings of the observed and computed responses for such data sets using a free-space model is rather difficult. Inversion of such data sets using a model in a free-space can yields larger size (strike length and depth extent) and smaller depth and conductance of the conducting body. Optimizing the effect of conducting host in each time channel together with the model parameters yields reliable estimates of the model parameters as well as proper fitting between the observed and computed responses. The procedure is based on the widely accepted qualitative assumption that the observed TEM response is the sum of the response due to the conducting target and the response due to the surroundings. It is also assumed that observations at a particular time channel (e.g. 4.5 ms) are affected by the same amount. Noise-free and noisy synthetic data affected by the conductivities of the surroundings are inverted. Efficacy of the approach is also demonstrated by three field examples measured in different geological conditions. The approach could also be used to invert transient EM responses using a variety of free-space formulations (e.g. a conducting plate in the layered Earth, a conducting spherical body in a conducting surrounding or layered Earth, etc.), where ever such assumptions are valid. Session 10 : other contributions Presentation: Oral or poster