| ERTLab | >  Command Reference for ERTLab

Command Reference for ERTLab

| << | >> |

ERTLab employs the following commands. These may be entered directly in the configuration settings displayed in both the inversion and forward workspaces, through an input file, or by using the menu system. 

Mesh settings

Type of mesh
Only mesh type 0 is available.

Interior Grid Element X Size (m)
Nominal X element size. If the mesh is adapted to electrode positions then this will vary modestly.

Interior Grid Element Y Size (m)
Nominal Y element size. If the mesh is adapted to electrode positions then this will vary modestly.

Interior Grid Element Z Size (m)
Nominal Z element size. If the mesh is adapted to electrode positions then this will vary modestly.
 
Mesh Minimum X Distance (m)
Nominal Z element size. If the mesh is adapted to electrode positions then this will vary modestly.

Mesh Maximum X Distance (m)
Nominal largest extent of the mesh in the X direction (value in meters).

Mesh Minimum Y Distance (m)
Nominal smallest extent of the mesh in the Y direction (value in meters). 

Mesh Maximum Y Distance (m)
Nominal largest extent of the mesh in the Y direction (value in meters). 

Minimum Z Value (m)
Nominal smallest extent of the mesh in the Z direction (value in meters).

Maximum Z Value (m)
Nominal largest extent of the mesh in the Z direction (value in meters).

Grid Right (+X) Pads (interior X grid units)
Pad elements that define the background region of the mesh in the positive X direction. For each number given, a single pad element is appended starting at the specified edge of the foreground region. Each subsequent number represents the ultimate size of the background region in this direction in terms of the equivalent number of interior mesh elements for that direction. The size of any single pad element is determined by subtracting the given value from the previous value and multiplying by the interior grid element size for the given direction. Default values are:  1 2 4 8 16 24 32 40 48 56 64; which specifies 11 pad elements whose sizes in meters are 1, 1, 2, 4, 8, 8, 8, 8, 8, 8 and 8 times the interior grid element size.

Grid Left  (-X) Pads (interior X grid units)
Pad elements that define the background region of the mesh in the negative X direction. For each number given, a single pad element is appended starting at the specified edge of the foreground region. Each subsequent number represents the ultimate size of the background region in this direction in terms of the equivalent number of interior mesh elements for that direction. The size of any single pad element is determined by subtracting the given value from the previous value and multiplying by the interior grid element size for the given direction. Default values are:  1 2 4 8 16 24 32 40 48 56 64; which specifies 11 pad elements whose sizes in meters are 1, 1, 2, 4, 8, 8, 8, 8, 8, 8 and 8 times the interior grid element size.

Grid Back  (-Y) Pads (interior Y grid units)
Pad elements that define the background region of the mesh in the positive Y direction. For each number given, a single pad element is appended starting at the specified edge of the foreground region. Each subsequent number represents the ultimate size of the background region in this direction in terms of the equivalent number of interior mesh elements for that direction. The size of any single pad element is determined by subtracting the given value from the previous value and multiplying by the interior grid element size for the given direction. Default values are:  1 2 4 8 16 24 32 40 48 56 64; which specifies 11 pad elements whose sizes in meters are 1, 1, 2, 4, 8, 8, 8, 8, 8, 8 and 8 times the interior grid element size.

Grid Front (+Y) Pads (interior Y grid units)
Pad elements that define the background region of the mesh in the negative Y direction. For each number given, a single pad element is appended starting at the specified edge of the foreground region. Each subsequent number represents the ultimate size of the background region in this direction in terms of the equivalent number of interior mesh elements for that direction. The size of any single pad element is determined by subtracting the given value from the previous value and multiplying by the interior grid element size for the given direction. Default values are:  1 2 4 8 16 24 32 40 48 56 64; which specifies 11 pad elements whose sizes in meters are 1, 1, 2, 4, 8, 8, 8, 8, 8, 8 and 8 times the interior grid element size.

Bottom (-Z) Pads (interior Z grid units)
Pad elements that define the background region of the mesh in the negative Z direction. For each number given, a single pad element is appended starting at the specified edge of the foreground region. Each subsequent number represents the ultimate size of the background region in this direction in terms of the equivalent number of interior mesh elements for that direction. The size of any single pad element is determined by subtracting the given value from the previous value and multiplying by the interior grid element size for the given direction. Default values are:  1 2 4 8 16 24 32 40 48 56 64; which specifies 11 pad elements whose sizes in meters are 1, 1, 2, 4, 8, 8, 8, 8, 8, 8 and 8 times the interior grid element size. Note that pad elements are not required in the positive Z direction because the boundary is always Neuman, i.e. no current flow.

Flat grid -1; Grid with topography -2
Use 1 for a mesh without topographic distortion, 2 to automatically adapt the surface (+Z) to the topographic relief specified either in the electrode ZTerrain or using a separate topography file.

Add electrodes Z terrain to topography: Yes-1; No-2
Use 1 to add the ZTerrain information from the electrode definition to the list of topographic points.

Update elevation for Surface electrodes: Yes-1; No-2
Use 1 to move Z coordinate of the surface electrodes to the topographic surface that was specified by importing a topography file. (Note that X and Y coordinates of the electrodes must be in the same coordinate system as the topography file).

Adapt grid to electrodes: Yes-1; No-2
Use 1 to snap (move) the nearest mesh node to the electrode location. Electrodes must fall on mesh nodes. If 2 is used then any electrodes that do not fall on nodes will be moved to the nearest node. Adapting the mesh is generally more accurate than moving the electrodes.

Starting model

Type of resistivity model
Only model type 0 is available.

Background Resistivity (ohm m)
The value entered here will be assigned to each element of the starting mesh, except when anomalies are supplied then those values will overwrite the background value supplied here.

Number of Resistivity Anomalies
The number of resistivity anomalies that will be employed in the starting model. Anomalies are used for forward modeling, and occaisionally to provide a more accurate starting guess for inversion.

Res. Anomalies: -X,+X  -Y,+Y  -Z,+Z  Resistivity
The location of a rectilinear resistivity anomaly. Seven values are expected, the negative and positive X extents (in meters), the negative and positive Y extents (in meters), the negative and positive Z extents (in meters) and the resistivity of the block (in ohm-m). More than one anomaly may be specified. Each subsequently specified anomaly overwrites the previous one if overlap occurs. Shapes other than rectilinear must be built using a series of these rectilinear anomalies.

Type of IP model
Only model type 0 is available.

Background Ip Value
The value entered here will be assigned to each element of the starting mesh, except when anomalies are supplied then those values will overwrite the background value supplied here.

Number of IP Anomalies
The number of IP anomalies that will be employed in the starting model. Anomalies are used for forward modeling, and occaisionally to provide a more accurate starting guess for inversion.

IP Anomalies: -X,+X  -Y,+Y  -Z,+Z  IP
The location of a rectilinear IP anomaly. Seven values are expected, the negative and positive X extents (in meters), the negative and positive Y extents (in meters), the negative and positive Z extents (in meters) and the IP of the block (in chargeability). More than one anomaly may be specified. Each subsequently specified anomaly overwrites the previous one if overlap occurs. Shapes other than rectilinear must be built using a series of these rectilinear anomalies.

IP Scale Factor (1000 for Chargebility)
A scale factor may be applied to IP data to convert from alternative units.

Add anomalies to flat grid: Yes-1; No-2
A value of 1 will add any anomalies (both resistivity and IP) to a flat, pre-topographic distortion mesh. Adaptations to a topographic surface (if used) are distributed throughout the mesh to minimize element distortion. Since any anomalies applied must completely fill the elements they affect, the rectilinearly defined anomalies will necessarily be distorted in the final mesh. This distortion is minimized if the anomalies are applied to the flat grid before the mesh is adapted to topography.

Forward solver parameters

[-/+X;-/+Y;-Z] Boundaries (0-Dir, 1-Mixed, 2-Neu)
Used to set the type of boundary conditions for each face of the model (edge of the mesh) except plus Z.  The plus Z direction is the earth’s surface and is always 2 (Neuman or no-flow).  –X and –Y indicate the first elements of the mesh in the X and Y directions, +X and +Y indicate the last mesh elements in the X and Y directions. Defaults are mixed (1) everywhere. Dirichlet or  zero voltage boundaries (0) may be used to represent a grounded electrode; and Neuman  or non-flow boundaries (2) may be used to represent an electrically isolated area such as a water tank or similar laboratory studies.

Omega: Used for SSOR precond

This parameter is used to adjust the convergence rate of the forward model. It controls the error reduction factor. Allowable values are 1.0 to 2.0.

Forward Solver Tolerance

This parameter indicates the stop criteria for the forward model. The smaller the value the more accurate the forward model, but longer processing time is required.

Maximum Number Forward Iterations
Indicates the maximum number of iterations for the forward model that will be performed. The larger the number the slower the solution.

Skip TX electrodes with no RX-1; Calc. All -2
A value of 1 will skip calculations for electrodes used only as transmitters. This may be useful to speed calcuation times when remote electrodes, which are typically only used as transmitters, are employed.

Model IP Data: No-1; Yes-2
A value of 2 will cause IP data to be modeled (either forward or inverse). This is performed after resistivity data are modeled.

Inversion parameters

Simple Inversion: Yes-1; No-2
ERTLab inversion strategy employs trials to determine the optimal roughness parameter that should be used for each outer iteration. Since this strategy results in significantly longer calculation times for each outer iteration for which it is employed, the Simple Inversion switch allows the user to turn the trials off for all but the first outer iteration.  If simple inversion is used then the roughness parameter used for subsequent outer iterations is the previous roughness parameter divided by the user controlled multiplier.

Maximum Number Inversion Iterations-Rho
This is the maximum number of outer iterations the user will allow the program to perform during inversion of resistivity data. Fifteen is usually a good starting point. The code will usually converge in fewer iterations.

Maximum Number Inversion Iterations-IP
This is the maximum number of outer iterations the user will allow the program to perform during inversion of IP data. Resistivity inversion happens before IP inversion. They do not happen concurrently. Ten is usually a good starting point. The code will usually converge in fewer iterations.

Maximum Internal Inverse PCG Iterations-Rho
This is the maximum number of inner iterations the user will allow the program to perform during inversion of resistivity data. Twelve is usually a good starting point.

Maximum Internal Inverse PCG Iterations-IP
This is the maximum number of inner iterations the user will allow the program to perform during inversion of IP data. Twenty is usually a good starting point.

Tolerance for Inverse PCG iterations-Rho
This parameter indicates the stop criteria for the resistivity inversion. Bigger  numbers  result in  earlier stopping.   Allowable values are .1 to .001.

Tolerance for Inverse PCG iterations-IP
This parameter indicates the stop criteria for the IP inversion. Bigger  numbers  result in  earlier stopping.  Allowable values are .1 to .001.

Factor <1 for choosing optimal Roughness Factor-Rho
This factor controls the identification of the optimal roughness parameter.

Factor <1 for choosing optimal Roughness Factor-IP
This factor controls the identification of the optimal roughness parameter

Multiplier for Changing Roughness Factor-Rho
This factor controls the degree of change between roughness parameters when stepping to the next outer iteration during simple inversion

Multiplier for Changing Roughness Factor-IP
This factor controls the degree of change between roughness parameters when stepping to the next outer iteration during simple inversion.

Initial Roughness Factor-Rho
This is the initial roughness parameter for the first outer iteration trial.

Initial Roughness Factor-IP
This is the initial roughness parameter for the first outer iteration trial.

Constant Value for Param X Weights
Used to control layering, factors smaller than 1 will increase layering in the X direction. Factors greater than 1 will increase smoothness.

Constant Value for Param Y Weights
Used to control layering, factors smaller than 1 will increase layering in the Y direction. Factors greater than 1 will increase smoothness.

Constant Value for Param Z Weights
Used to control layering, factors smaller than 1 will increase layering in the Z direction. Factors greater than 1 will increase smoothness. The default is to impose layering in the Z direction.

Noise

Rho Data Percent Errors (%)
This parameter controls the acceptable percent difference between the data and the model (the inversion will converge to this error level). The value entered here should reflect the estimated degree of data repeatability (i.e. the actual noise in the field data), which is easily obtained if reciprocal data were collected. This term works in conjunction with the constant error term to determine convergence.

Rho Data Constant Error Term
This parameter indicates the acceptable absolute difference between the data and the model that is acceptable (the inversion will converge to this error level). The value entered here should reflect the estimated lower noise threshold of the instrument used to collect the data. This term works in conjunction with the percent error term to determine convergence.

IP Data Percent Errors (%)
This parameter controls the acceptable percent difference between the data and the model (the inversion will converge to this error level). The value entered here should reflect the estimated degree of data repeatability (i.e. the actual noise in the field data), which is easily obtained if reciprocal data were collected. This term works in conjunction with the constant error term to determine convergence.

IP Data Constant Error Term
This parameter indicates the acceptable absolute difference between the data and the model that is acceptable (the inversion will converge to this error level). The value entered here should reflect the estimated lower noise threshold of the instrument used to collect the data. This term works in conjunction with the percent error term to determine convergence.

Use robust inversion (data errors reweight)-1; No -2
A value of 1 sets use of robust inversion. This strategy allows the code to ignore data outliers (those data with the worst model fit) so they are no longer important to the solution. Except in the most rigorously controlled field situations this strategy should be used. When inverting synthetic or forward data this option should be turned off.  Even when robust is used  it is possible that  the inversion will not converge. This may happen when all of the data have approximately the same deviations and no outliers can be identified.

Number of reweight iterations-Rho
This parameter controls the number of trials that may be attempted when searching for an optimal roughness parameter.

Number of reweight iterations-IP
This parameter controls the number of trials that may be attempted when searching for an optimal roughness parameter.


Saving and Plotting parameters

Plot bounds: -X +X -Y +Y -Z +Z
These parameters control the extents of the output model. Typically only the foreground region is output. However for thin models (e.g. 2D models), the extents output should be widened. The units are in meters.

Save inverted model for each iteration: Yes-1; No-2
A value of 1 causes a model to be output (written to disk) after each outer iteration.


 



|© 2006 Geostudi Astier srl and Multi-Phase Technologies LLC|