GP33B-1249
Electromagnetic imaging of hydrates and accretionary structure at Hydrate Ridge, Oregon using 2.5 D model studies
We have carried out model studies of controlled source electromagnetic (CSEM) data collected at Hydrate Ridge, Oregon, over the Cascadia accretionary margin. This project was designed to image near-surface gas hydrate using CSEM, but we also collected broad-band marine MT data. A 2D inversion of the MT data provides an image of folding associated with the accretionary complex and is sensitive to conductive fluids within the accretionary prism. In order to improve our understanding of the distribution of hydrate in the shallow section we need similar inversion tools for CSEM data. We compared the use of finite element (Li and Key, 2007) and finite difference (Abubakar et al.,2006) CSEM 2.5 D modeling codes (3D source, 2D model). In particular, we modeled bathymetric changes of up to 800 m along a 20 km, E-W profile across the ridge. The finite element and finite difference codes accommodate bathymetry in different ways: the finite difference approach uses a rectangular grid that extends throughout the entire model space, whereas the finite element approach uses triangular elements that can be refined around any complex geometry without propagating through the model. We compare the efficiency and accuracy of these two codes. Fixed resistivity models incorporating bathymetry were then run to create look-up tables and thus transform our electric field data into apparent resistivities. We compare pseudosections generated this way with those derived using flat-earth 1D calculations. We also attempted to invert the field data using the finite difference 2.5 D pixel inversion code. Inversion results give better depth constraints on resistivity structure, which can then be compared with 2D MT inversion results and seismic data. http://marineemlab.ucsd.edu/
GP33B-1250
Invariant-based imaging and interpretation of magnetotelluric measurements over an area and along a profile in SW-Transdanubia, Hungary
We applied various imaging and interpretation approaches to a magnetotelluric data set in SW-Transdanubia, Hungary: a) data from about 300 MT soundings, carried out over an area of about 40 km x 25 km, and b) data from about 100 MT soundings, measured along the CELEBRATION 007 profile with a step distance of appr. 2 km. Different authors had recommended different solutions for the invariant presentation of the magnetotelluric results: the so-called SM-, WAL-, phase tensor invariants, and all decomposition techniques also belong to this category. In this paper we give a systematic summary about the results obtained with all these parameters in a wide period range. In all approaches it should be kept in mind, that the magnetotelluric impedance tensor contains seven (not more and not less) real data of about the resitivity distribution of the subsurface. While we have got a mixed experience with some fashionable invariant parameters calculated from field data, some phase- and the real tensor invariants have proven to be useful. They indicated deep structures at unambigously shorter periods than the common resistivity or imaginary tensor-based data. It is also interesting that the phase tensor suggests less thicker deep structures than expected from geology. SW-Transdanubia is crossed by two main tectonic lines: the Mid-Hungarian- and the Balaton lines. The Mid-Hungarian line, although it separates the two basic units of the Pannonian Basin (Alcapa and the Tisia), seems to have weeker and more intermittent magnetotelluric features than the very robust Balaton line.
GP33B-1251
Marine Electromagnetic Imaging of the Catalina Crater
In May 2006, we conducted a combined magnetotelluric (MT) and controlled source electromagnetic (CSEM) survey of the Catalina Crater, located in the Inner Continental Borderland off the coast of southern California. The Catalina crater is visible in seafloor bathymetry as a circular feature 30 km in diameter, with a morphology similar to a large crater, including a raised outer rim, ring moat and a central peak. Seismic, gravity and magnetic data show similarities to known impact craters, yet enough ambiguity exists that an alternative volcanic origin cannot be ruled out. Our marine electromagnetic survey was designed to map electrical conductivity from the seafloor to 30 km depth in order to provide further constraints on the origin of the crater. We deployed a 38-kilometer line of 12 seafloor MT/CSEM receivers, extending over the eastern three quarters of the crater and out over the trough directly to the east. CSEM transmissions were performed by deep-towing the Scripps Undersea EM Source Instrument (SUESI) along a 30 km tow line over the receivers. SUESI transmitted with a 95.2 kAm source dipole moment and used a complex binary waveform consisting of fundamental frequencies of 0.1, 1, and 10~Hz. The MT data provide constraints on conductivity variations to 30 km depth and are augmented with CSEM data for imaging conductivity from the seabed to a few kilometers depth. Good quality MT responses were obtained in the period band of 5-5000 seconds and show large conductivity variations across the survey line. As a first cut interpretation of the CSEM data, we derived apparent resistivities from changes in amplitude of the recorded signal with range, and mapped the apparent resistivities into a resistivity pseudosection for each frequency. The pseudosections slice through the eastern half of the crater and the adjacent trough, providing a first-order glimpse of the lateral variations of conductivity. Our electromagnetic study of the crater will provide new insights into the crater origin by allowing us to distinguish between conductive porous sediments and resistive volcanics and basement formations.
GP33B-1252
Comparison of 1- and 2-D Regularized MT Inversions
We measure the success of a general numerical optimization scheme for solving the nonlinear inverse problem of electromagnetic induction (MT) in two dimensions by comparison with a one-dimensional analytic solution. The general, large-scale method treats the differential equations as constraints and the observations together with the 2-norm of the (positive) conductivity as a penalty function. The one-dimensional comparison problem is solved by means of the Euler-Lagrange equations with a positivity condition for a single frequency observation, but the solution can then be easily extended to many frequencies. Having exact analytic solutions for a simple system provides a valuable check on complex general-purpose codes.
GP33B-1253
Magnetotelluric Evidence of Thick-skinned Thrusting in the Southern Tien Shan, Kyrgyzstan and China?
A profile of broadband MT stations was acquired in 2005 in association with a planned 500 km long seismic reflection line shot in 2007. This profile extended from south of Kashgar, China, to just south of Bishkek, Kyrgyzstan and consisted of 73 stations spanning the northern part of the Tarim Basin and most of the Tien Shan mountains in Kyrgyzstan. Here, we report on a model derived from the southernmost 40 broadband and long period stations that coincides with a seismic line shot by Chinese geophysicists. Broadband data confirm earlier results that the conductive sediments in the Tarim Basin extend to almost 10 km depth. Most intriguing is an antiformal feature that spans depths from 10 to 40 km whose crest lies beneath the Kashgar anticline. This antiformal feature lies beneath detachment faults at depths of 6 to 8 km that have been identified from surface geology and seismic reflection data by other authors. It also lies beneath the deepest structures visible in the 2007 reflection line. While it is possible that the structure is a basement relict from an earlier tectonic phase, the spatial association with late Cenozoic deformation supports an interpretation that it is young. If so, then the lithosphere beneath the shallower detachment faults is also deforming.
GP33B-1254
A Modular System for EM Inversion: Implementation of NLCG for 3-D Magnetotelluric Data
We are developing a general modular system for gradient based inversion of electromagnetic (EM) data. The inversion code has been designed using an object oriented approach, with the highest levels of functionality independent of problem specifics. In particular, top level modules manipulate abstract data objects (such as data vectors, model parameters, EM field solutions) and methods (e.g., data functionals, solvers), to implement a range of model and data space gradient based search algorithms. Previously, the system has been used as a test-bed for experimentation with new inversion algorithms, using the 2D magnetotelluric (MT) problem as a test case. Here we describe progress on using this general modular system to develop a non-linear conjugate gradients inversion for 3D MT data. The kernel of this new inversion is a 3D staggered-grid finite difference forward solver. The very specific interfaces of the forward code, both with respect to model parameters and source/boundary conditions, facilitate incorporation into the general modular system. The full adjoint (including sensitivity to boundary conditions) for the solver was developed as an intrinsic part of the forward code, making application in a gradient based inversion scheme such as non-linear conjugate gradients (NLCG) particularly straightforward. Development of such an efficient, ultimately parallel, gradient-based inversion has been motivated by the newly available, intrinsically 3D MT data sets such as those collected in the framework of EarthScope. We expect that the modular approach will simplify development and testing of alternative parameterization and regularization schemes, more readily allow inversion of novel data types (e.g., inter-station or magnetic gradient transfer functions), and accelerate implementation of new more efficient search algorithms.
GP33B-1255
Preliminary Report on Seafloor Electromagnetic Observations off Tottori in the Sea of Japan
Seafloor electromagnetic observations were made off Tottori in the Sea of Japan together with MT measurements in the land region, in order to investigate the rsistivity structure in the San-in region, the Tottori and surrounding region in the northern part of Chugoku district, southwestern Japan, where epicenters of microearthquakes are remarkably distributing within a line belt with a width of about 4-9 km along the coastal line of the Sea of Japan. The depths of the hypocenters are located up to about 10km depths. In the seismic belt, several large earthquakes of M6.2-7.4 took place in 1943, 1983 and 2000. Moreover, quaternary volcanoes, such as Daisen volvano and Oginosen volcano are also located in the seismic belt. Wide-band magnetotelluirc (MT) observations have been made along survey profiles of almost N-S direction in the San-in region since 1998, to investigate heterogeneity in the crustal electrical resistivity structure. Through those wide-band MT measurements we found the low resistive region beneath seismogenic zone of the high seismicity belt on each MT profile line, and that the upper resistive crust corresponds to the seismogenic zone in the Tottori and northern Hyogo region. The low resistive region found along each wide-band MT profile seems to form a conductive zone extending in the almost E-W direction beneath the seismic belt extending in the almost same direction of the conductor. This result strongly suggests the existence of crustal fluid beneath the seismogenic zone in the focal area. The survey lines should be extended toward the Sea of Japan and longer period MT data on the land area should be obtained, in order to investigate deeper resistivity structure and clarify the relationship between subducting Philippine Sea plate and the deeper resistivity structure beneath the San-in region. Therefore, we carried out seafloor and longer period MT surveys on land . It is important to analyze both data sets to obtain high accuracy resistivity strucure. OBEMs and OBEs were deployed and recovered in 2006 and 2007 to obtain seafloor geoelectromagnetic data beneath the off-Tottori of the Sea of Japan in the San-in region. High-quality MT data from the seafloor array, together with those of three additional land sites in the San-in region, is analyzing now. We will show the outlines of the observation and the preliminary results along the first survey line in presentation.
GP33B-1256
Challenge to image the stagnant slab beneath the Philippine Sea by seafloor electromagnetic survey
We have run a seafloor electromagnetic survey project in the Philippine Sea in order to image the deep mantle slab stagnating in the transition zone and surrounding mantle in three dimensions. Seafloor observations at every 500 km or so is necessary to resolve the geometry of the slab because existing data sets are based on the observations by land geomagnetic stations and submarine cables, which are distributed coarsely and unevenly. Although it is difficult to establish a bunch of nearly permanent observation stations at seafloor, iterative maneuver observations using ocean bottom electromagnetometers (OBEMs) can acquire the data required to probe down to the mantle transition zone. The project iterates one-year-long deployment three times. Earthquake Research Institute, University of Tokyo and Institute for Research on Earth Evolution (IFREE), Japan Agency for Marine-Earth Science and Technology (JAMSTEC) have resourced the project with the OBEMs. In the first phase, we deployed 11 OBEMs in October, 2005 and recovered all of them successfully in November, 2006. R/V Kairei of JAMSTEC was utilized for both cruises. In the second cruise, we deployed another 12 OBEMs and started the second phase. The recovery of the OBEMs and the third initiation will be done in November, 2007. The quality of the first phase data is quite good except for the electric field at one site. The MT responses are estimated at 10 sites in the period range of from about 300 to 60,000 seconds. The geomagnetic transfer functions are also estimated at all the 11 sites in the range of about 300 to 1,000,000 seconds. The responses at the period longer than 100,000 seconds are somewhat inaccurate and the improvement is expected by further data collection in the second and third phases. These responses will be analyzed together with the responses which were obtained by past experiments in the Philippine Sea. The features of the responses may be classified by basins composing the Philippine Sea plate. Some of these features are explained by land-ocean distribution and seafloor topography. Three-dimensional forward modelling study is now on going to examine effects of the surface heterogeneities and some simple patterns of sub-surface heterogeneities. These results and preliminary report on the second phase will be presented in the meeting.
GP33B-1257
Magnetotelluric Surveys Using Near and Very far Remote References
Magnetotelluric (MT) soundings have been made on two traverse lines to the north and west of the Century Zinc mine in NW Queensland, Australia, in order to image the intermediate to deep structure in the region. The purpose of the MT surveys lies in delineating the extension of the Termite Range Fault (TRF), a major structure on the Lawn Hill Platform running through the Century Zinc mine with north-west strike direction. During the survey, we operated the remote reference (RR) site in Gregory Downs, which is roughly 80 km apart from the field site. High quality MT signals have been obtained using remote reference techniques. Besides the RR site in Australia, a permanent monitoring site in Esashi, Japan, which is approximately 4600km away, had been operating during the whole period of the survey simultaneously. MT data were acquired in the field location simultaneously together with remote data at remote stations with the help of GPS time synchronization. Long period pulsations observed in this survey generally showed dominant Hx polarization than Hy. We also observed high coherency between MT signal at the very-far remote reference site and local site for two frequency ranges; around 1 Hz and 10-2 Hz, especially in Hx component. We tried to extract MT impedances using RR in Japan and compared the results from two RR's. The results shows that apparent resistivities and phases processed with two remote reference magnetic data are compatible.
GP33B-1258
Insights into signal and noise in long period MT data from multi-station processing of EarthScope arrays
During 2006-2007 long period (10-10000 s) MT data were collected at approximately 100 sites covering Washington and Oregon as part of the EarthScope transportable array experiment. Site occupations in a series of overlapping two dimensional arrays, and the relatively large spacing (75 km) make this a unique data set for studies of signal and noise characteristics in long period MT data, and to explore novel data processing and interpretation strategies. I will present results from applying multiple station array processing methods based on principal component (SDM eigenvector) analysis to a series of small overlapping arrays, each consisting of 5-10 sites operating for 10-15 days. Data quality was exceptional in 2006, and poorer in 2007. Causes of this decline in quality will be explored. Very short spatial scales associated with PC3 pulsations are clearly evident in data from all arrays near a period of 30 s; source effects in MT impedances (but not always in vertical field TFs) are generally negligible at this period. At longer periods (> 1000 s) gradients in the magnetic field, and associated "normal Hz" can be associated with smaller significant coherent signal components. In addition, there are large scale coherent electric fields clearly evident in essentially all arrays at the longest periods. These electric fields scale with the plane wave impedance, and probably reflect galvanic current flows induced at some distance from the local site, and hence not always coherent with local magnetic fields. We will also present results of initial efforts to compute 3D spatial gradient transfer functions, and to construct hypothetical events of the full EM surface fields.