Synthetic Data
==============

The section shows some synthetic studies made using MARE2DEM. Synthetic
forward studies can be highly useful for determining whether or not a
given structure generates a detectable MT or CSEM response and for
seeking out the optimal frequency range for maximizing the anomalous
signals.  Once you establish whether or not a structure can generate a
detectable EM response signal, the next test is to carry out synthetic
inversion studies to see if the data can resolve the structure, since
non-uniqueness in EM inversion means data *sensitivity* does not
necessarily translate into data *resolution*. Synthetic inversion
studies can also be used to optimize receiver and transmitter layouts,
frequency ranges, impacts of various relative and absolute noise levels,
etc.



Crosswell EM
------------


.. figure::  _static/images/examples/croswell_fwd.png
    :width: 60%
    
    Forward model used for crosswell EM study. 400 Hz amplitude and
    phase data were generated for 49 vertical magnetic dipole
    transmitters (circles) with 49 vertical magnetic dipole receivers
    (diamonds). 1% noise was added to the forward responses to create
    synthetic data for inversion. Note the topography and undulating
    model structure shown here is the actual polygonal model layering
    precisely modeled in MARE2DEM using adpative finite element meshes.

.. figure::  _static/images/examples/croswell_inv.png
    :width: 60%

    Inversion fitting the data to RMS 1.0. Note how the conductive layering and
    prism body are well recovered but the resistive body is not recovered, as 
    expected for inductive magnetic transmitters and receivers.


.. figure::  _static/images/examples/croswell_inv_grid.png
    :width: 60%
    
    Inversion model now shown with the quadrilateral free parameter grid boundaries 
    outlined in black.
    
.. figure::  _static/images/examples/croswell_datafits.png
    :width: 100%
    
    Synthetic crosswell EM data and model fits shown in matrix form.
    
Subglacial MT Imaging of Groundwater
------------------------------------

Modified from: Key, K., & Siegfried, M. R. (2017), The feasibility of
imaging subglacial hydrology beneath ice streams with ground-based
electromagnetics. Journal of Glaciology, 331, 1–17. DOI:
`10.1017/jog.2017.36 <http://doi.org/10.1017/jog.2017.36>`_. This
conceptual model study paper helped lead to funding for the `SALSA EM
survey in Antarctica
<https://emlab.ldeo.columbia.edu/index.php/projects/subglacial-em-
mapping/>`_.
    
.. figure::  _static/images/examples/GZ_complex_wet.0.resistivity_fwd.png
    :width: 100%
    
    Resistivity forward model for subglacial groundwater at the
    grounding line of the Whillans Ice Stream, West Antarctica, where it
    becomes the Ross Ice Shelf. Blue regions show seawater saturated
    sediments and isolated saline aquifers.

    
.. figure::  _static/images/examples/GZ_inv.27.resistivity.png
    :width: 100%
    
    MARE2DEM inversion of synthetic MT data generated for the grounding
    zone model above. The ice layer (100,000 ohm-m) was held as fixed
    structure with known resistivity. Black lines show the structural
    boundaries from the true model. The dark conductive features are well
    resolved by the MT array, showing how MT data can be useful for
    mapping subglacial groundwater and brines beneath ~700 m of ice.