13:45 〜 15:15
[SEM14-P07] Comparison of MT responses in spherical Earth and in Cartesian system by different map projections
Magnetotelluric (MT) is a passive exploration method which usually assumes the Earth surface is flat. Map projections enable us to model MT responses obtained in a region. However, any map projection introduces geometric deformation at locations away from the map center, which may cause inaccuracy in estimating MT responses. Here, theoretical impedance tensors in spherical Earth and those in several Cartesian systems by different map projections are compared and quantitatively discuss deviations in resulting impedance tensors.
We tested a regional model consisting of a 4-km-thick inhomogeneous shell representing land-sea contrast and a 1-km-thick inhomogeneous shell representing sediment thickness variation. Below two surface thin shells, 1-D structure is assumed. We choose six sites on the Philippine Sea plate and one site on the western edge of the Pacific plate where ocean bottom electromagnetometers (OBEMs) were installed (e.g. Baba et al., 2010). We first evaluate the impedance tensor in a target region (40 degree x 40 degree) centered at (25 °N, 135° E) in spherical Earth model using a global forward code by Uyeshima and Schultz (2000). The model spatial resolution is 1degree by 1 degree in the target region but is degraded outside. Two perpendicular magnetic dipole (axial and equatorial) fields at the center of the target region are given as the external source fields. Then, we estimate impedance tensors in cartesian coordinate system based on different map projection methods by using ModEM code (Egbert and Kelbert, 2012). Model space is a 4000 km x 4000 km square region and spatial resolution is 100 km. Here Cylindrical, azimuthal, conic as well as thematical map projections are considered. The impedance tensor elements estimated at each site location in XY coordinate system are rotated to theta-phi direction to be compared with those estimated in spherical Earth.
Additionally, we tried another spherical model by rotating the coordinate axes so as the center of the study region to be on the equator. In this situation, deformations of geometry in the northern and southern regions of the study region become symmetric after map projection, and therefore better agreement is obtained between results in spherical and Cartesian models.
References
Uyeshima, M. and Schultz, A. Geoelectromagnetic induction in a heterogeneous sphere: a new 3-D forward solver using a staggered-grid finite difference method, Geophys. J. Int., 2000, 140, 636-650.
Egbert G D, Kelbert A. Computational recipes for electromagnetic inverse problems[J]. Geophysical Journal International, 2012, 189(1): 251-267.
Baba K, Utada H, Goto T, et al. Electrical conductivity imaging of the Philippine Sea upper mantle using seafloor magnetotelluric data. Physics of the Earth and Planetary Interiors, 2010, 183(1-2): 44-62.
We tested a regional model consisting of a 4-km-thick inhomogeneous shell representing land-sea contrast and a 1-km-thick inhomogeneous shell representing sediment thickness variation. Below two surface thin shells, 1-D structure is assumed. We choose six sites on the Philippine Sea plate and one site on the western edge of the Pacific plate where ocean bottom electromagnetometers (OBEMs) were installed (e.g. Baba et al., 2010). We first evaluate the impedance tensor in a target region (40 degree x 40 degree) centered at (25 °N, 135° E) in spherical Earth model using a global forward code by Uyeshima and Schultz (2000). The model spatial resolution is 1degree by 1 degree in the target region but is degraded outside. Two perpendicular magnetic dipole (axial and equatorial) fields at the center of the target region are given as the external source fields. Then, we estimate impedance tensors in cartesian coordinate system based on different map projection methods by using ModEM code (Egbert and Kelbert, 2012). Model space is a 4000 km x 4000 km square region and spatial resolution is 100 km. Here Cylindrical, azimuthal, conic as well as thematical map projections are considered. The impedance tensor elements estimated at each site location in XY coordinate system are rotated to theta-phi direction to be compared with those estimated in spherical Earth.
Additionally, we tried another spherical model by rotating the coordinate axes so as the center of the study region to be on the equator. In this situation, deformations of geometry in the northern and southern regions of the study region become symmetric after map projection, and therefore better agreement is obtained between results in spherical and Cartesian models.
References
Uyeshima, M. and Schultz, A. Geoelectromagnetic induction in a heterogeneous sphere: a new 3-D forward solver using a staggered-grid finite difference method, Geophys. J. Int., 2000, 140, 636-650.
Egbert G D, Kelbert A. Computational recipes for electromagnetic inverse problems[J]. Geophysical Journal International, 2012, 189(1): 251-267.
Baba K, Utada H, Goto T, et al. Electrical conductivity imaging of the Philippine Sea upper mantle using seafloor magnetotelluric data. Physics of the Earth and Planetary Interiors, 2010, 183(1-2): 44-62.