2D model maker for DCIP2D forward modelling

Related documentation:
| DCIP2D user interface | data viewer | model maker | model viewer | tutorial |

This page outlines how to use the program dcip2d-model-maker for building 2d resistivity and chargeability models in the UBC-GIF rectangular mesh format.

Installation

The following programs are required to build 2D models, run a forward calculation and view the result as a pseudosection. Inversion is not discussed in this document.
 

NOTE REGARDING SAVING WORK:
Most UBC-GIF programs save results to fixed file names. Consequently each new forward modelling or inversion run should be started from a new directory. Otherwise previous files will be overwritten.

Getting started

If "dcip2d-model-maker.exe" is run from the command line (or by double clicking it's icon) it will open directly with a blank window. There are no command line parameters for this program.

When you start dcip2d-model-maker.exe the dialogue to the right will appear. These parameters define the model space. Spatial dimensions are in any units you like, but the resulting resistivities etc. will be in the same units. For example if you enter dimensions in feet the resistivities will be in Ohm-feet rather than Ohm-meters. Consequently it is advisable to use metres.

• x1 = left edge in model units
• x2 = right edge in model units
• depth = in model units
• log of the background conductivity in milliSiemens per metre (mS/m).
• background chargeability in your chosen units (see below).

Upon clicking the OK button, The model building window appears as follows. The program starts in the "DC" mode, in which you build 2D models of variations in subsurface resistivity. See below for notes on working in the "IP" mode.

Zones of varying resistivity are placed within this model domain by drawing rectangles with the mouse. Click with the left mouse button at any corner of your desired new block, then drag the mouse to the opposite corner. Upon releasing the mouse button a dialogue appears with current parameters for the new block. This is where the block's new resistivity is entered. The spatial coordinates can also be adjusted at this point. Click OK to accept the parameters as defined.

This programs features are all available using the buttons on the tool bar. There are also menus selections that duplicate these features, as well as permit the opening of previously saved models.

Modifying model blocks

• The move and size options allow you to move or resize the block using the mouse.
• The properties option permits precise adjustment of spatial coordinates and physical properties. Note that you can work in resistivity or conductivity, and using linear or log values, depending upon how the "units" buttons  have been selected.
• Delete is self explanitory, and to back changes the order in which multiple blocks are drawn on the screen.
• The copy option makes an identical block of chargeable material with a default chargeability of 0.5 in dimensionless units. Adjust this block's physical properties in the IP mode (see below).
• The shear option permits you to change the block so there is a slope to the vertical edges.

Including topography

To add topography click the "topography" button . Then the mouse cursor will become a small + . Add nodes for topography by double clicking at suitable locations to simulate the required surface topography. Once there is topography, point to a topography node with the right mouse button (while in the topography mode) to delete nodes or remove topography alltogether. To return to adjusting buried block parameters click the "topography" button again.

To have topography extend beyond the model region across the left and right padding cells (described in the next section), double click a point outside the model region. Two examples are given below.

Three topography nodes - one just outside the left edge of the region of interest, one just outside the right edge of the region of interest, and one at the peak near the center of the model.

Three topography nodes - one just inside the left edge of the region of interest, one just inside the right edge of the region of interest, and one at the peak near the center of the model.

Discretizing the model

Before running a forward calculation your model must be discretized onto a mesh using the UBC-GIF format. This is a very important step and the result of discretizing the model you have defined will significantly affect the data that result from running a forward calculation. This calculation is performed on a discrete model, so you must be sure you are satisfied with the way the program has converted your subsurface model into the discrete domain.  Carry out the discretization by clicking the mesh button . If you have not already saved your model you will be reminded at this point. You are permitted to specify the name of the model to be saved. See the note regarding saving to new directories above.

Parameters in the "Discretize Model" dialogue should be self explanatory, except that the "aspect ratio" means the ratio of width to depth of the cells in the top few layers.  An aspect ratio of 2:1 is normal for these top few rows, with aspect ratio reduced to 1:1 for the padding cells below the region of interest. Note that it is not recommended to request fewer than 2 cells per electrode, but more may help produce smoother results. Of course, more cells will result in lengthier calculations.

Upon accepting the discretization parameters, the model that will be used for forward calculations is displayed in a separate model display window.  Also, the discretization, as well as a sketch of the survey geometry, is overlaid on the model in the modelling window.  For the model that will be used in forward calculations, the region of interest will be augmented with "padding cells" which are required to ensure that boundary conditions at the edges of the domain modelled by inversion are well controlled. For example, the two figures in the section on topography above show how a region of interest ranging from (x1, x2) = (0, 200), with depth extent of 100 is augmented to include padding cells to the left, right, and underneath of the region of interest.

Using the model display window (which is the same as that used to display inversion results), observe how your buried blocks and topography have been converted to rectangular cells. Use the mesh-view button in the model viewing window to see how the mesh has been imposed upon the subsurface features.  There are more instructions on using dcip2d-model-viewer in a separate help file. If the discretization you see is not suitable, go back to the modelling window and adjust the model. Otherwise you are ready to perform the forward calculations.  (The model display can be invoked anytime there is a model available by clicking the  button.)

Wenner and Schlumberger arrays are not included in the electrode geometries. These can be simulated by generating a pole-pole data set, then working with the resulting data file to generate data that would arise from a different survey configuration by subtraction of appropriate pairs of pole-pole data. Recall that the research priority at the UBC-GIF is inversion theory and methodology, and graphical user interface programs that accompany forward and inversion codes are constructed to support research rather than to satisfy the needs of third parties.

Running a forward calculation

Once you are satisfied with the discretization that has been applied you are ready to perform a forward calculation. Click the forward modelling button .  A command line ("DOS") window will appear to indicate progress of the calculation.

When the forward calculation is finished, the command line window will disappear, you can examine the resulting data as a pseudosection by clicking the  button. Details on using dcip2d-data-viewer are provided in a separate document.

Working in IP mode

So far we have only worked with resistivity models. By clicking the "IP" button, you can add subsurface blocks representing variations in chargeability. Notice that the resistivity structures are outlined in dashed lines while working in the chargeability mode.  Forward modelling of chargeability data is performed using the same sequence of steps as for resistivity, but you must be in the IP mode (with the "IP" button down) throughout.

Currently, the program can work only with theoretical dimensionless intrinsic chargeability, so values must be between 0.0 and 1.0. See the UBC-GIF website (http://www.eos.ubc.ca/research/ubcgif/documentation/dcipdocs/intro.html) for details of this definition for chargeability. To understand why these non-dimensional units are used, it should be recalled that inthe development of the IP inversion algorithms, the assumption was made that the IP response is the apparent chargeability obtained from an idealized time domain experiment.  Using the figure to the right, the apparent cheargeability is defined as the ratio of the secondary potential _s to the total potential measured _n which was measured just before the current is cut off. This is a dimensionless number and usually has a range of 0 to 0.5.  In reality _s and _n are not generally recorded directly with the field acquisition system. In order to use UBC-GIF programs so that the output is a true "chargeability" defined by Siegel, 1959, it is necessary to convert collected IP data into the form assumed by these algorithms. An approximate conversion can be performed using the following approximate rules of thumb:

NOTE:   The resistivity model should be correct before IP forward calculations are performed. You can adjust the resistivity model by clicking the "DC" mode button - then the chargeability blocks are outlined in dashed lines. 

References:

1. Van Voorhis,G.D.,  P.H. Nelson, and T.L. Drake, 1973, "Complex resistivity spectra of porphyry copper mineralization", Geophysics, Vol. 38, pg. 984.
2. Siegel, H.O., 1959, "Mathematical formulation and type curves for induced polarization", Geophysics Vol. 24, pg. 547.