Model Targets - Interactive Mode

Use the Interactive Mode option (geosoft.uxo.gxnet.dll(Geosoft.GX.UXO.UxaLaunchAddReview;Run)*) to review the fitted parameters for the existing targets, refit the parameters for the existing targets, or model the newly added targets.

Modelling Parameters - Dynamic Survey Mode dialog options

General Information

Sensor type

Select the sensor type from the list of available types:

  • EM61
  • EM61MK2
  • EM63
  • EM61MK2-Array
  • Mag

Script Parameter: UXANALYZE.SENSORNAME

Database

Select the method of collecting site data for standard sensors:

  • Select Chronological to view the line profile of your targets.
  • Select Non chronological if no line profile is required.

Script Parameter: UXANALYZE.CHRONOLOGICAL [1 - chronological; 0 - non-chronological]

Subset database

Select whether to update or use the existing subset database.

  • The Update option extracts data from the site database and saves the data to the subset database.
  • The Use existing option uses the data directly from the subset database if it exists. If it does not exist, it will extract the data from the site database.

Script Parameter: UXANALYZE.UPDATESUBSETDB [1 - update; 0 - use existing]

Site Database

Name

Specify the survey database name.

Script Parameter: UXANALYZE.GDB

Sensor channel

Select the sensor channel from the list of channels in the servey database.

Script Parameter: UXANALYZE.SENCHAN

Sensor height - Variable channel

Select the sensor height channel from the list of channels in the site database.

Script Parameter: UXANALYZE.ALTCHAN

Sensor height - Constant

Specify a constant height applied to all the site data.

Script Parameter: UXANALYZE.CONSTHEIGHT

Pitch channel

Select the pitch channel from the list of channels in the site database.

  • This entry is only visible when the sensor type is EM.
  • Script Parameter: UXANALYZE.PITCH

    Roll channel

    Select the roll channel from the list of channels in the site database.

  • This entry is only visible when the sensor type is EM.
  • Script Parameter: UXANALYZE.ROLL

    Heading channel

    Select the heading (yaw) channel from the list of channels in the site database.

  • This entry is only visible when the sensor type is EM.
  • Script Parameter: UXANALYZE.HEADING

    Coil channel

    Select the coil channel from the list of channels in the site database.

  • This entry is only visible when the sensor type is EM61MK2-Array.
  • Script Parameter: UXANALYZE.COILCHAN

    Target Database

    Name

    Specify the target database name.

    Script Parameter: UXANALYZE.TARGETGDB

    Group

    Select the target group from the list of groups in the target database to invert.

    Script Parameter: UXANALYZE.TARGETGROUP

    ID channel

    Select the ID channel from the list of channels in the target database.

    Script Parameter: UXANALYZE.TARGETID

    Start ID

    Specify the starting ID value. Only the targets with the ID equal to or greater than this value will be processed. 

    Script Parameter: UXANALYZE.STARTID

    End ID

    Specify the ending ID value. Only the targets with the ID equal to or less than this value will be processed.

    Script Parameter: UXANALYZE.ENDID

    Mask channel

    Select the target database mask channel to filter out specific targets. Only the targets with a mask (channel value) of 1 will be processed. If the value is set to 0 or dummy (*), the corresponding target will be ignored.

    If the field is left blank, no mask will be applied.

    Script Parameter: UXANALYZE.TARGETMASK

    Sensor Parameters Information; Mag - IGRF

    Inclination

    Specify the inclination value.
    Magnetic parameters can be calculated via the IGRF GX.

    Script Parameter: UXANALYZE.INCLINATION

    Declination

    Specify the declination value.
    Magnetic parameters can be calculated via the IGRF GX.

    Script Parameter: UXANALYZE.DECLINATION

    Total field

    Specify the total field value.
    Magnetic parameters can be calculated via the IGRF GX.

    Script Parameter: UXANALYZE.FIELDSTRENGTH

    Sensor Parameters Information; EM61

    Coil size

    Select the coil size from the available options: 1x1 m, 0.5x1 m, or 1x0.5 m. Default selection is 0.5x1 m.

    Script Parameter: UXANALYZE.COILSIZE

    Mode

    Select either Mode D or Mode 4. Default selection is D.

  • This entry is only visible for EM61-MK2.
  • Script Parameter: UXANALYZE.MODE

    Channels to fit

    Select the gate(s) to be used in the modelling process.

    Script Parameter: UXANALYZE.CHANNELSTOFIT

    Coil separation

    Specify the coil separation.

  • This entry is only visible for EM61.
  • Script Parameter: UXANALYZE.COILSEP

    Sensor Parameters Information; EM61MK2-Array - Coil Geometry

    This tab only appears when the sensor type is a EM61 sensor array. The Coil Geometry tab tabulates the coils' locations.

    Configuration

    The configuration contains the number and location of each sensor in the sensor array, which are read from SensorTableLand.xml.

    Refer to Application tips below for more information about the Sensor table file.

    Set

    When selected, the updated configuration is saved into a copy of SensorTableLand.xml located in the %USERPROFILE%\Documents\Geosoft\Desktop Applications \etc folder. In the future execution of this dialog, this new configuration is loaded instead of the original one.

    Factory setting

    When selected, the original configuration is loaded into the dialog.

    Preview

    When selected, the sensor array configuration is shown in a schematic map.

    [ More ]

    Maximum gap in units of average sample spacing

    Specify the maximum gap value to select the profile line. Default is 5.

    This entry is only visible when a chronological database is selected.

    Script Parameter: UXANALYZE.NGAPS

    Maximum deviation angle along the profile

    Specify the maximum deviation angle to select the profile line. Default is 90.

    This entry is only visible when a chronological database is selected.

    Script Parameter: UXANALYZE.DEVANGLE

    Application Notes

    The user will select Add/Review Targets from the menu, fill the various entries from the menu, fill the various entries in the Modelling Parameters dialog, and click OK. The tool prompts the user with the entries to create the site map if it doesn’t exist. The map name is generated based on the site database name and the first sensor channel:

    SITE_DB_@@SENSOR_CHAN@@.map

    The map will be opened, and the Add/Review Targets interface will show up. At this time, control is passed to the user, who can select on the map an existing target or a new one.
    The whole map is shown as a thumbnail at top of the dialog, with a red rectangle showing how much of the area is covered by the site map. The size of the rectangle is determined by the Add/Review window sizevalue defined by the user in the Settings GX. Moving the rectangle using the mouse changes the view in the site map.
    When the user clicks on the map, the tool looks in the target database to find an existing target in that location. If a target is found, its information is shown in the Selected Point Information section of the GUI. However, if there is no existing target in that location, the Target ID value will change to <new target> and all the buttons at the bottom of the GUI are disabled except Add and Help.
    When Add is clicked, a new target will be added to the target database, its ID will appear in the Target ID field, and the disabled buttons will be enabled.
    You are able to use the Fit button to calculate the fitting parameters regardless of whether it is a new or an existing target. All the other buttons can also be used for an existing target.

    Prepare your target data base with the X & Y channels of each potential target, an ID channel with a unique number for each target, and a mask channel turned to 1 for all the targets to be Modelled.

    The X & Y channels should be of type "double".

    The ID channel should be of type "string", short, or long.

    The Mask channel should be of type "short"

    For fast access, the process sorts the survey data by X then Y into one big bin. Then for each target it extracts a window of data centered on the target, and of the specified window size. This data is saved in a subset database, the name of which is build as:

    <Survey database>_<Target database>_<Target group>.gdb

    For each target, the extracted data is saved in a random group by Target ID number.   In addition to this window, the profile closest to the target is also extracted. All the targets for which the mask is turned on are then fitted.  The calculated target data is saved in the subset database along side of the observed data.  The target database is updated with the fitted parameters listed below. 

    Magnetic Target Parameters

    Fit_error: Error flag

         0 - No error

         1 - Anomaly not strong enough to fit

         2 - Not enough data points selected for Modelling

         3 - Zero RMS noise

        -3 - Reached maximum number of iterations

         4 - Memory allocation error

         5 - Square root of negative number encountered when fitting target

         6 - Division by zero

         7 - Matrix inversion problem     

         8 - Parameter values changed too much relative to initial estimates    

         9 - Fit process did not converge

       10 - Problem encountered in lamda optimization loop

       11 - Data has not been Modelled

     999 - Target has been removed

    Fit_coh: A measure of how well the Modelled data fits the measured data; specifically, the correlation coefficient squared. A high fit coherence does not guarantee that the fit results are accurate – only that the match between Modelled and measured data is high. Multiple solutions may exist. The fit quality also depends on the number of data samples being passed to the fitting algorithm. The fewer the samples the easier it is to get a high fit coherence ( dimensionless)

    Fit_depth: Fitted depth of the target from the surface (survey distance units)

    Fit_x: Fitted X location of the target (survey distance units)

    Fit_y: Fitted Y location of the target (survey distance units)

    Fit_dec: Fitted declination (degrees)

    Fit_inc: Fitted inclination (degrees)  

    Fit_MagneticMoment: magnetic dipole moment, a measure of the of the magnetic source ( Amp-m2)

    Fit_size: Empirically calculated diameter of the object (distance units). The size estimate output from the magnetic data are a function of the object orientation with respect to the earth’s field. The standard deviation of the size estimate is approximately 30% of the ordnance diameter. The apparent size is inferred from the radius of steel sphere which has the same magnetic dipole as the target. The overall EMI response is proportional to the volume of a compact steel target. The target size calculated from the EM data is determined by adding the three polarizabilities (betas) and applying a proportionality constant so the average size calculated using the first time gate data approximately scales with the UXO caliber. As with the size calculated with magnetic data there is variability in the size output from the EM inversion. In general, the size output from magnetic data is more robust than the size using EM data for common targets. tip that the size calculation has not been calibrated for the different time gates therefore for later time gates. The “fit_size” parameter should be used as a relative indicator of size and not as an absolute indicator of UXO caliber.

    Fit_Solid angle: The angle between the earth&#39;s magnetic field and the inverted source dipole ( degrees)

    Index2SubsetGDB: The group number in the subset database holding the corresponding data

    Comments: Initially this string is of zero length - the user may populate it with target related information. 

    EM61 Target Parameters

    Fit_error: Error flag

       0- No error

       1 - Anomaly not strong enough to fit

       2 - Not enough data points selected for Modelling

       3 - Zero RMS noise -3 - Reached maximum number of iterations

       4 - Memory allocation error for target

       5 - Square root of negative number encountered when fitting target

       6 - Division by zero

       7 - Matrix inversion problem

       8 - Parameter values changed too much relative to initial estimates

       9 - Fit process did not converge

    Fit_coh: A measure of how well the Modelled data fits the measured data; specifically, the correlation coefficient squared. Ideally a fit coefficient of 0.995 or greater is desired for EM data in order to have confidence in the shape information (expressed by the betas – see definition below). However, if only size and depth are of interest, then a lower fit coherence (i.e. ~0.98) may be acceptable. A high fit coherence does not guarantee that the fit results are accurate – only that the match between Modelled and measured data is high. Multiple solutions may exist. The fit quality also depends on the number of data samples being passed to the fitting algorithm. The fewer the samples the easier it is to get a high fit coherenceFit_depth: Fitted depth from surface (dimensionless)

    Fit_depth: Fitted depth of the target from the surface (survey distance units)

    Fit_x: Fitted X location of the target (survey distance units)

    Fit_y: Fitted Y location of the target (survey distance units)

    Index2SubsetGDB: The group number in the subset database holding the corresponding data

    Comments: Initially this string is of zero length - the user may populate it with target related information

    ModelingChannels: in the case of EM61MKII the channels used to derive the fitted parameters. tip that if multiple channels are Modelled, multiple sets of fitted parameters will result. The later fitted target parameters are appended to the bottom of the target group.

    Fit_b1, b2, b3: Three response coefficients, referred to as betas, that represent the eigenvalues of the magnetic polarizability tensor. The relative size of the coefficients provides information about the shape of the object. For cylindrical, metallic, and isolated objects, we expect b1>b2≈b2. For spherical objects, b1≈b2≈b3. They can be thought of as components along the three orthogonal axes of the induced dipole.

    Fit_theta: Inclination of the source dipole (degrees)

    Fit_chi2: Goodness of fit measure. ∑((y-yfit)/sdev)2

    Fit_phi: Declination of the source dipole (degrees)

    Fit_psi: Rotation of the source dipole (degrees)

    This methodology has been provided by SAIC Inc. 

    SensorTableLand.xml Lookup Table

    The SensorTableLand.xml lookup table file contains various information, which is used to populate the entries in the dialogs and to define the fitting channels. The file is located in the "C:\Program Files\Geosoft\Desktop Applications \etc" folder.

    If you would like to customize the file, you should copy the file SensorTableLand.xml to the "%USERPROFILE%\Documents\Geosoft\Desktop Applications \etc" folder; otherwise, your changes may be lost during a software update.

    This tool is developed in partnership with Acorn Science and Innovation (AcornSI).

    *The GX tool will search in the "gx" folder. The GX.Net tools, however, are embedded in the Geosoft.uxo.gxnet.dll located in the bin folder. If running this GX interactively, bypassing the menu, first change the folder to point to the bin folder, then supply the GX.Net tool in the specified format.