Auxiliary material for Paper 2010GC003101 Upper mantle electrical resistivity structure beneath the central Mariana subduction system Tetsuo Matsuno Department of Applied Ocean Physics and Engineering, Woods Hole Oceanographic Institution, Woods Hole, Massachusetts, USA Nobukazu Seama Research Center for Inland Seas, Kobe University, Kobe, Japan Rob L. Evans Department of Geology and Geophysics, Woods Hole Oceanographic Institution, Woods Hole, Massachusetts, USA Alan D. Chave Department of Applied Ocean Physics and Engineering, Woods Hole Oceanographic Institution, Woods Hole, Massachusetts, USA Kiyoshi Baba Ocean Hemisphere Research Center, Earthquake Research Institute, University of Tokyo, Tokyo, Japan Antony White School of Chemistry, Physics and Earth Sciences, Flinders University, Adelaide, South Australia, Australia Deceased 22 June 2007 Tada-nori Goto Department of Civil and Earth Resources Engineering, Kyoto University, Kyoto, Japan Graham Heinson and Goran Boren School of Earth and Environmental Science, University of Adelaide, Adelaide, South Australia, Australia Asami Yoneda and Hisashi Utada Ocean Hemisphere Research Center, Earthquake Research Institute, University of Tokyo, Tokyo, Japan Matsuno, T., et al. (2010), Upper mantle electrical resistivity structure beneath the central Mariana subduction system, Geochem. Geophys. Geosyst., 11, Q09003, doi:10.1029/2010GC003101. Introduction: The auxiliary materials comprise five supplementary figures in order to present results of topographic correction by using two different sturcutres of the 1-D structure (Table 2) and the optimal 2-D structure (Figure 6c), a 2-D model including TE mode apparent resistivity in inversion, 2D models excluding some MT response functions in inversion, and a sensistivity map of the optimal 2-D structure (Figure 6c). 1. 2010gc003101-fs01.pdf Figure S1. MT apparent resistivity topographically-corrected by using the 1-D structure of Table 2 and the 2-D structure of Figure 6c. White circles indicate the responses by using the 1-D structure, and black circles indicate the responses by using the 2-D structure. Error bars indicate double-sided one-standard deviation. Site names and site numbers shown on the upper left in each box are those used in 2-D inversions (refer to Table 1). 2. 2010gc003101-fs02.pdf Figure S2. MT phase topographically-corrected by using the 1-D structure of Table 2 and the 2-D structure of Figure 6c. White circles indicate the responses by using the 1-D structure, and black circles indicate the responses by using the 2-D structure. Error bars indicate double-sided one-standard deviation. Site names and site numbers shown on the upper left in each box are those used in 2-D inversions (refer to Table 1). 3. 2010gc003101-fs03.pdf Figure S3. Inversion models by using the TE mode apparent resistivity. Inversion settings such as smoothing parameters and model mesh are the same as those for Figure 6a. The upper figure is a model using all available data for 2-D inversion listed in Table 1, the middle figure is a model without data at two sites (site number 10 and 20), and the lower figure is a ratio of the two models. A total RMS misfit of the upper model is 1.634, and that of the middle model is 1.616. 4. 2010gc003101-fs04.pdf Figure S4. Inversion models without the TE mode apparent resistivity. Inversion settings such as smoothing parameter and model mesh are the same as those for Figure 6c. The upper figure is a model using all available data for 2-D inversion listed in Table 1 (this is the same figure as Figure 6c), the middle figure is the model without two sites data (site number 10 and 20), and the lower figure is a ratio of the two models. A total RMS misfit of the upper model is 1.122 and that of the middle model is 1.128. 5. 2010gc003101-fs05.pdf Figure S5. A sensitivity map for Figure 6c. The sensitivity value is defined as the square of diagonal of At V-1A, where A is the Jacobian matrix and V is the data error covariance matrix, normalized by each numerical mesh dimensions and a maximum sensitivity value.