Two-dimensional Electrical Resistivity Model from San Pablo Bay, California
Model and response of 2D electrical resistivity model
Dates
Publication Date
2020-04-10
Start Date
2015-10-05
End Date
2015-10-09
Citation
Peacock, J.R., Ponce, D.A., Watt, J.T., and K.M., Denton, 2020, Magnetotelluric data from San Pablo Bay, California: U.S. Geological Survey data release, https://doi.org/10.5066/P96T27UM.
Summary
Included here are the model, data, and reponse files used to create the 2D electrical resistivity model. Plots of the data and model response are provided as well as one of the model. The data, model, mesh, regularization, and response files are provided in Occam2D format. See https://marineemlab.ucsd.edu/Projects/Occam/2DMT/index.html for details on file format. This data product includes Occam2D data, model, response, and mesh files such that anyone can reproduce the model results. To produce the 2D model the data are rotated to principle strike direction estimated from the phase tensor azimuth, which was nominaly geomagnetic north. The station locations are projected onto a profile line N66E. Bad data points are removed by hand. [...]
Summary
Included here are the model, data, and reponse files used to create the 2D electrical resistivity model. Plots of the data and model response are provided as well as one of the model. The data, model, mesh, regularization, and response files are provided in Occam2D format. See https://marineemlab.ucsd.edu/Projects/Occam/2DMT/index.html for details on file format. This data product includes Occam2D data, model, response, and mesh files such that anyone can reproduce the model results. To produce the 2D model the data are rotated to principle strike direction estimated from the phase tensor azimuth, which was nominaly geomagnetic north. The station locations are projected onto a profile line N66E. Bad data points are removed by hand. Only the TM and induction vectors are inverted to remove 3D effects that might occur from being so close to the coast. The starting model was a homogeneous half-space of 10 Ohm-m. The inversion ran for 20 iterations where the 10th iteration was deemed to be the optimum based on an L-curve of roughness versus root-mean-squared error. The final normalize root-mean-squared error is 1.98 using errors of 10% for the apparent resistivity, 1.5 degrees for the phase and 0.2 for the induction vectors. The final relative roughness is 174.