Inputs

Every input is optional. A method's _model is the 3D volume and its _data is the measured dataset for the same method.

Input	Formats	Loaded as
MT_model	ModEM/WS .rho	Resistivity volume
MT_data	ModEM .dat	Georeferences the mesh; Show Data draws the site locations
gravity_model	.xyz, .vox	Density volume, displayed in g/cc
gravity_data	.xyz	Show Data draws the survey points at their own elevation
magnetic_model	.xyz, .vox	Susceptibility volume
magnetic_data	.xyz	Show Data draws the survey points
seismic_data	SEG-Y	Seismic curtain and model drape
shapefiles	.shp	Linework, each entry with its own color, width and alpha

What you get

What you get depends on what you configure:

Configured	Result
Nothing	An empty 3D scene, sized to the shapefiles or seismic line if there are any
MT_model only	The model in its own local coordinates; georeferenced overlays are left out
MT_model + MT_data	Georeferenced into coordinate.target_crs
A gravity/magnetic model with an MT model	Resampled onto the MT cells, so the properties compare cell for cell
A gravity/magnetic model without one	Gridded on axes taken from its own coordinates

Reading point files

.xyz models and surveys are longitude latitude elevation value records, or easting/northing in any projected CRS. They are reprojected into the project CRS and, for models, resampled onto the scene's cells.

Source CRS detection

A file that states no CRS has one worked out for it. Degrees and projected metres never overlap in practice: longitude/latitude is bounded by ±180/±90, while a metric CRS puts survey data tens to hundreds of kilometres from its false origin. A file is therefore read as WGS84 lon/lat or as already being in the project CRS accordingly. When both columns fall inside ±90 and the lon/lat order is ambiguous, the project CRS's declared area of use decides which column is longitude, so latitude-first files land in the right place too. .prj files and ModEM origins are still read from the file itself.